# RoundCut: full content > RoundCut is a free image-tools site. 21 live tools today across 29 locales (3 flagship tools, 12 image-format conversion pairs, 4 more single-purpose tools, 2 family hubs that route by category). This file carries every live tool page, every format-conversion page and every institutional page in full. Different tools use different paths — full-browser, hybrid client-preview / server-encode, or cloud inference — and each entry says which. Generated 2026-09-16 from the site's content collections. Each entry below opens with `# ` and a `Source:` line carrying its canonical URL, so any slice of this file traces back to the page it came from. The short curated index is at https://roundcut.app/llms.txt. The blog lives at https://roundcut.app/blog/ (English) and https://roundcut.app/blog/pt/ (Portuguese), and indexes itself at https://roundcut.app/blog/llms.txt and https://roundcut.app/blog/pt/llms.txt — each entry there links the post as markdown. The whole blog as one raw corpus: https://roundcut.app/llms-blog-en.txt and https://roundcut.app/llms-blog-pt.txt. ## What RoundCut is A free image-tools site without accounts, signups, watermarks, or per-image / per-day limits. The principle is simple: build the most complete tool reasonable, then describe exactly how it works. Where the right architecture is fully client-side, we do it that way. Where a server-side encoder produces a materially better file, we say so plainly. We never claim "no upload" for a tool that uploads. ## Principles - **Truthful claims.** Marketing copy mirrors what the code actually does. We don't claim "no upload" for tools that upload. - **No accounts, no signups, no watermarks, no per-image / per-day limits.** Tools are funded by us, with ads on a small subset of pages. - **Privacy means different things on different tools.** Circle Crop never uploads. Compress sends the file for the final encode but holds it in a tenant-isolated cache and deletes it after a short TTL. Background Remover sends the file to Cloudflare edge GPUs for inference. Convert pairs run entirely in the browser. Each tool's entry describes its own privacy contract. - **No fake numbers.** No fabricated star ratings, user counts, or download counts. ## Locales Site is available in 29 languages. This file is the English edition. Each tool page below also exists at `/<tool>/<locale>/` (e.g. `/compress/pt/`) and each institutional page at `/<locale>/<page>/` (e.g. `/pt/about/`) — the two shapes differ, and only one applies per page type. Locale list: https://roundcut.app/llms.txt. ## Canonical URL https://roundcut.app/ --- # Remove Background Source: https://roundcut.app/background-remover/ > Drop a JPG, PNG, or WebP and the background remover lifts the subject out for you, handing back a clean cutout on a transparent PNG, a lighter WebP, or a white-background JPEG. **Architecture:** by default, BiRefNet (a 200MB segmentation model) running on Cloudflare edge GPUs through a Cloudflare Worker. When the cloud path is unavailable (network down, file > limit, service degraded), the tool falls back to an in-browser engine (onnxruntime-web + an open isnet segmentation model, served from our own origin) — slower (5–60s on mid-range mobile vs ~5s on cloud), same quality. The cloud encode never trains on your file; the file is held in tenant-isolated R2 storage for a short TTL then removed. **Inputs:** JPG, PNG, WebP. **Outputs:** PNG with transparent background. Up to 4096 × 4096 pixels for the cloud path; the browser fallback is bounded by your device's memory. A background remover lifts your subject out of a photo and clears everything behind it, so you get a clean cutout in seconds. Drop a JPG, PNG, or WebP and the tool isolates the main subject, then hands it back on a transparent background that drops onto any color, photo, or layout without a pale box around it. It reads people, products, pets, food, and most everyday things, even against a busy scene. Want a plain backdrop instead? Save as JPEG to place the subject on a clean white background, or keep a transparent PNG to paste it anywhere. Either way the cutout comes back at the same size you started with, and you can zoom right in to check the hair and edges before you save. It works the same on your phone and your computer. For how your photo is handled, see the privacy page. ## What it does - Remove the background from a photo automatically - Clean cutout with a transparent background - Holds fine hair and fur edges - Output as transparent PNG, lighter WebP, or white-background JPEG - Cutout returns at your original resolution - Zoom in to inspect the edges before download - Input support for JPG, JPEG, PNG and WebP - Works on people, products, pets, food and everyday objects ## How to use it 1. **Choose your image** — Tap the upload area or drag a JPG, PNG, or WebP photo onto the page. 2. **Let the tool lift the subject** — The background remover isolates the main subject and clears everything behind it in a few seconds. 3. **Check the edges** — Zoom in on the cutout to make sure the hair, fingers, or product rim look clean. 4. **Pick how to save it** — Choose a transparent PNG, a lighter WebP, or a white-background JPEG. 5. **Download your cutout** — Save the finished cutout and drop it onto any color, scene, or layout. ## How to remove the background from your image Drop your photo and the background remover gets to work right away. It isolates the main subject and clears everything behind it. You get back a clean cutout with a transparent background. It sits on any color, scene, or layout with no white box around it. There is nothing to trace by hand. The tool reads people, products, pets, food, furniture, and most everyday things. It hands you the result in a few seconds. Check the cutout, pick how you want to save it, and download. The same steps remove the bg from a product shot, a portrait, or a busy holiday photo. ## Choose how to save your cutout Once the background is gone, you choose the file that fits where the cutout is going. Pick a transparent PNG to paste the subject onto another image or a colored layout. PNG can make a background transparent and works everywhere. Pick WebP for a lighter file that looks the same on the web or by email. When the place you are using it cannot read a transparent edge, save as JPEG instead. The cutout then sits on a clean white background. When in doubt, stay with PNG. Need a different file type for the cutout? You can <a href="/convert/#tool">convert it</a> without leaving RoundCut. ## What the background remover handles well The tool was built for real photos, not just flat studio shots. It reads a subject against a busy room, a street, or a crowd. Portraits, product photos on any surface, pets, food, cars, and everyday objects all come out clean. Fine hair and fur edges hold up well too. A few subjects are genuinely harder. Clear glass with liquid behind it, a mirror, thin backlit fabric, or a pale subject on a pale background can trip it up. There the tool has to guess where the subject ends, and it can be off. Expect a little touch-up. When an edge looks rough, you can <a href="/#tool">crop it round</a> to tidy the frame. ## Your cutout keeps its full size The cutout comes back at the same width and height you started with. A large photo gives you a large cutout, with no shrinking and no surprise downscale. That keeps the subject sharp in a tiny avatar or a full-width banner alike. When you save as WebP or JPEG, a quality control lets you trade a little detail for a smaller file. It stays out of the way when you keep the transparent PNG. Before you download, zoom right into the edge to check the hair, the fingers, or the rim of a product. If your photo started small, <a href="/upscale-2x/#tool">add resolution first</a> so the cutout reads sharp. ## What happens to your photo Your photo is handled only for the few seconds it takes to lift the subject out, and then it is cleared. Nothing is kept, nothing is logged, and nothing is passed to anyone else. There is no tracking on the result you download. The cutout is yours to use however you like, on a profile, a store listing, a deck, or a poster. If you want the full detail of how files move through the tool and how long anything is held, the privacy page lays it out plainly. The point is simple: drop a photo, get a clean cutout back, and move on without leaving a trail behind you. ## Put your cutout to work A clean cutout is the starting point for a lot of small jobs. Drop it onto a brand color for a product card, layer it into a thumbnail, or place it over a fresh scene for a post. When the file feels heavy for the web, <a href="/compress/#tool">compress the cutout</a> down to a smaller size with no visible drop in quality. When it needs to fit an exact frame, <a href="/crop-image/#tool">crop it</a> to the shape you want. Each of those steps lives right here. You go from a raw photo to a finished, ready-to-post image in one place. No juggling apps, no uploading the same picture twice. ## Questions **What can I do with the cutout once the background is gone?** Almost anything that needs a subject without its surroundings. Place it on a brand color for a product card, layer it into a thumbnail or a poster, build a logo or a sticker, or set it over a fresh scene for a post. Because the edge is transparent, the subject blends into whatever sits behind it, with no pale box giving away that it was cut from another photo. **Which image formats can I upload?** You can upload JPG, JPEG, PNG, and WebP, the formats the background remover reads with full color. If you have a HEIC photo straight from an iPhone, just drop it in and the tool converts it to JPG for you. Very large files may take a little longer, but the tool still returns a clean cutout from the same photo you started with. **Which output formats can I download?** Three. A transparent PNG keeps the see-through edge and is the safe pick when you plan to paste the cutout onto something else. A WebP keeps your transparent edge too, in a lighter file for the web. A JPEG places the cutout on a clean white background, which is handy when the next tool cannot read a transparent edge. PNG covers most cases. **Can it remove a white or solid background?** Yes. A plain white or solid-color backdrop is one of the easiest cases, since the subject stands clearly apart from it. Drop the photo in and the tool clears the white background and hands back a transparent cutout. If you would rather keep a clean white backdrop for a listing, save as JPEG instead and the subject sits on solid white. **Does it handle hair, fur, and fine edges?** Usually well. The background remover was trained on real photos, so thin hair and soft fur edges hold up cleanly on most portraits and pets. The hard cases are see-through subjects like clear glass, fine mesh, or a thin backlit fabric, where the tool has to decide what counts as the subject. For those, expect a little manual cleanup after you download the cutout. **Does it work on my phone?** Yes. The background remover opens in any modern phone browser, so you can lift a subject out of a photo straight from your camera roll without installing anything. The steps are the same as on a computer: drop the photo, wait a few seconds, check the cutout, and save. For HEIC photos from an iPhone, the tool converts them to JPG for you, so just upload. **Is it good for product photos?** Very. A product photo reads best when nothing competes with the item, so lifting it off your kitchen table or desk makes it pop. Drop your shot in, get a clean cutout, and the same item looks sharp on a website, a catalog, or an ad. You can repeat it for a whole range and every shot will share the same tidy look. **Can I use the cutout for a logo or a sticker?** Yes. A transparent cutout is exactly what a logo, a sticker, or a watermark needs, since it carries no boxy backdrop into the design. Drop the subject onto a label, a packaging mockup, or a chat sticker pack and it blends in. If the shape needs to be round or square for the spot it is going, trim it to that frame after you save. ## The details ### Why a transparent cutout drops onto any background When the background is gone and you save as a transparent PNG or WebP, the empty space around your subject stays truly empty instead of turning white. That is what lets the cutout sit on a brand color, a photo, or a busy layout without a pale box framing it. The same file works on a dark site header and a light printed flyer, because the area around the subject simply shows whatever is behind it. If you ever need an opaque backdrop instead, a JPEG fills that space with white. Keeping the cutout transparent is almost always the more flexible choice, since you can always add a background later but you cannot easily remove one you baked in. ### A clean cutout for store listings and marketplaces Most marketplaces want a product to sit on a plain white or transparent backdrop so the item, not the room behind it, is what shoppers see. Lifting the subject out gives you exactly that: a tidy cutout you can place on white for a listing or keep transparent for a banner. Shoot the product on any surface you have, drop the photo in, and let the tool clear the clutter behind it. The result reads consistent across a grid of listings, which makes a small shop look far more put together. From there you can crop the cutout to the frame the platform asks for and compress it so the page stays fast to load. ### When to remove the background and when to crop instead Removing the background is the right move when the thing behind your subject is the problem, like a messy desk behind a headshot or a crowd behind a product. If the subject itself is fine and you only need a tighter frame or a round shape, a crop is faster and keeps the original photo untouched. Many photos benefit from both: clear the background first, then crop the clean cutout to a circle for an avatar or a rectangle for a thumbnail. Thinking about what is actually in the way, the backdrop or the framing, points you to the quicker path and saves you from over-editing a photo that only needed one small change. --- # Circle Crop Source: https://roundcut.app/ > A circle cropper for your profile picture, avatar, and round logo. Drop your image, frame the circle, and download with a transparent background. **Architecture:** runs 100% in your browser using Cropper.js plus the HTML5 Canvas API. Your file is decoded, cropped, and exported entirely on your device. You can confirm this in DevTools → Network: zero outbound image requests for this tool. **Inputs:** JPG, JPEG, PNG, WebP, GIF (first frame), AVIF, BMP, SVG, HEIC/HEIF if your browser can decode it. **Outputs:** PNG (transparent), WebP (transparent), AVIF (transparent), or JPEG (with solid background color of your choice). Up to 4096 × 4096 pixels. To circle crop image online, upload your photo and drag the round selection over the face, object, or logo. Move the image and resize the circle until you frame exactly the part you want to keep. Pick PNG, WebP, or AVIF so your round photo keeps a transparent background. Prefer a solid color behind it? Save as JPEG instead. The result is perfect as a profile picture, avatar, round logo, or team icon, and it drops cleanly onto any background without a white box around it. Need an exact size before you save? Resize your image first. And if the background is busy, you can remove it before you crop. It all works on desktop and mobile, and your circle crop image is ready to download in seconds. For how files are handled, see the privacy page. ## What it does - Adjustable circle crop for precise framing - Round profile photos, avatars, logos and team icons - Output as PNG, WebP, AVIF or JPEG - Transparent area outside the circle in PNG, WebP and AVIF - Custom background color for JPEG - Input support for JPG, PNG, WebP, AVIF, GIF, BMP and SVG - HEIC depending on browser support - Export at the original size, up to 4096 px ## How to use it 1. **Choose your image** — Tap the upload area or drag a JPG, PNG, WebP, AVIF, GIF, BMP or SVG file onto the page. 2. **Adjust the circle** — Move and resize the circular selection until the face, subject or logo is framed the way you want. 3. **Pick your format** — Choose a transparent format for a clear see-through edge, or JPEG with a background color you pick. 4. **Save the round image** — Check the preview and download the circle-cropped image in your chosen format. ## How to circle crop an image Drop your photo and the round selection opens right in the middle. The area outside is dimmed, so you can see what stays and what goes. Drag the circle until the face, product, or logo is framed the way you want. Adjust it from the corners, or pinch with two fingers on your phone. Check the preview and tap Save to download your image already round. The same steps let you circle crop image files from a square photo or a wider shot. You stay in control of the framing the whole time. There is nothing to install and nothing to learn. Just open the page and crop your photo on the spot. ## Choose the format for your round photo Want your round photo with a transparent background, so it sits on any color behind it? Pick PNG, WebP, or AVIF. PNG is the safe choice and works everywhere. WebP gives you a lighter file that looks the same. AVIF goes even smaller when you want to save space. And if the place you are using your image does not accept transparency, choose JPEG. Your round photo then sits on a solid color you pick yourself. When in doubt, stay with PNG, which covers the vast majority of cases. That way your round photo lands right, from a chat avatar to a logo on your site. ## A round profile picture for every social network Almost every social network shows your avatar inside a circle. Doing the circle crop yourself makes sure your face sits right, instead of letting the platform cut it any way it likes and clip your head. The trick is to leave a little room above the head and lift the eyes slightly above center. Then your profile picture looks balanced, from the big profile view down to the tiny feed thumbnail. The same trick works for your round logo, team photo, and server icon. If the background gets in the way of your photo, you can <a href="/background-remover/#tool">remove the background</a> before you come back and crop. ## The right size for your avatar When you circle crop image files for an avatar, you do not need to memorize any size. Each network shows your round photo at a different size: large on the profile and much smaller next to a comment. They all shrink the image for you to fit the screen. So the secret is simple: start from a sharp photo and leave the right gap around the face. If you only have a small version, you can <a href="/upscale-2x/#tool">upscale the resolution</a> before you crop. And when a specific case asks for an exact size, <a href="/resize/#tool">resize your image</a> first so your avatar always looks sharp and never blurry, even at the smallest size. ## A circle cropper that fits any image The circle cropper handles the everyday images you throw at it, from a phone snapshot to a high-resolution headshot or a wide group photo. To crop a circle in your image, drag the file onto the page, place the round selection over the part that matters, and download. You choose PNG, WebP, or AVIF for a transparent edge, or JPEG with a background color you pick. The round selection moves and scales on its own, so you frame the exact region you want, the same way on desktop and mobile. Export stays at high resolution, so your circle crop looks sharp from a tiny feed thumbnail up to a large profile view. ## What this tool does not do A few honest notes, so you do not waste your time. iPhone photos in the HEIC format open straight in Safari, and in other browsers the tool converts them to JPG for you automatically before you crop. An animated GIF becomes a round photo from the first frame only, so the animation does not carry over. And the crop is done by you, by hand, with nothing automatic deciding where to cut. The circle of your round photo sits exactly where you want it. Apart from those few cases, the circle cropper works with any everyday photo you have, from a portrait to a logo, and the control is all yours. Drop a photo, frame the circle, and your round crop is ready, with no surprise waiting at the Save step. ## Questions **How do I crop a photo into a circle?** Drop your photo onto the page and the round selection opens centered on it. Drag the circle to frame the face, product, or logo, adjust it from the corners or with a pinch on your phone, and tap Save. By default you get a round photo with a transparent background, ready to use. The whole circle crop takes just a few seconds on any everyday photo, and you can fine-tune the framing as much as you like before you download. **How do I crop a circle image from a square one?** Drag your square image onto the page and the circle opens taking up almost the whole photo. Reposition and adjust the circle to frame what matters, then click Save. To crop circle image files from a square, PNG, WebP, and AVIF keep the four corners transparent, so only the round part stays visible. If you would rather have your round photo on a solid color, just choose JPEG and set the color that fills the corners before you download. **What file formats are supported?** You can upload the everyday image types like JPG, PNG, WebP, and a few more. iPhone photos in the HEIC format open in Safari, and in Chrome, Firefox, or Edge the tool converts them to JPG for you automatically. The tool reads the real file type, so even a renamed file is recognized correctly. In the end, you download your round photo as PNG, WebP, or AVIF with a clear see-through edge, or as JPEG with a background color. **Will my cropped image have a transparent background?** Yes, when you choose PNG, WebP, or AVIF the area outside the circle comes out transparent. So your round photo drops onto any background color with no white square around it. When you open the file in a design app or upload it to a social network, that outside area shows as a checkerboard, which is just how transparency is displayed. If you prefer a solid background, pick JPEG and set the color that fills the corners. **Can I use this tool on my mobile device?** Yes, the circle cropper works in any modern browser on iPhone, iPad, and Android, with nothing to install. The handles respond to touch, and you use a two-finger pinch to move and size the circle right around the face. For a quick round photo on your phone, go with PNG or WebP, which are ready almost instantly even on simpler devices. The result is the same as on desktop, with your photo always staying with you. **Why crop an image into a circle?** Most social networks show your avatar inside a circular frame, so a square photo gets clipped at the edges in ways you do not control. Cropping the circle yourself decides exactly what stays inside the ring, which keeps the face centered and balanced. Beyond avatars, a round crop suits logos, team badges, sticker art, and any design where a circular shape reads cleaner than a square one. It is a small step that makes a profile picture look intentional rather than auto-cut. **How do I make a circle profile picture for Discord or LinkedIn?** Upload your photo, then drag and size the circle so the face sits in the middle with a little room above the head. For Discord and LinkedIn, start from a sharp, fairly large photo and let the platform shrink it to the avatar size. Leave a bit of breathing space around the face so it stays clear at the small sizes used next to messages and in the feed. Save as a transparent PNG and the round avatar drops cleanly onto the platform's background. **How are my files handled?** How your images are processed and whether anything is stored is described on the privacy page, which is the single place that stays current as the site evolves. The tool itself focuses on one job: opening your photo, letting you frame the circle, and exporting the round result in the format you choose. For the full, up-to-date detail on file handling, read the <a href="/privacy/">privacy page</a> before you rely on any specific behavior. ## The details ### Why your round photo sits on any background The real magic of a circle crop saved as PNG, WebP, or AVIF is that everything outside the circle truly disappears. In practice, your image lands clean on any background color, with none of that white square around it that gives away a cropped photo. That is why it looks right in the round frame on LinkedIn, Discord, or Instagram, and also when you place your round photo over other images in a design. When the spot needs a solid color behind it instead of transparency, just save as JPEG and pick the color. Everything else stays exactly as simple, and your round photo is ready to share wherever you need it. In short, that clean transparent edge is what makes a profile picture look professional on any network, with no leftover white box, just the circle and you inside it. ### Circle crop or rectangular crop: which to use A circle crop is the right call when the result lands in a round frame, like an avatar, a community icon, or a badge, where the corners would be hidden anyway. It keeps the subject centered in the ring and drops onto any background once you export with a transparent edge. A <a href="/crop-image/#tool">rectangular crop</a> makes more sense when you want to keep straight edges, trim to a banner, or set a specific width and height for a thumbnail or header. Many people do both: a rectangular trim first to remove clutter, then the round cut for the avatar. If you are unsure, start round for anything that becomes a profile picture and rectangular for anything that fills a box. ### Tips for a great round profile picture A few framing habits make a round photo read well at any size. Keep the eyes slightly above the center of the circle, since that is where attention lands first, and leave a little space above the head so the crop does not feel tight. Fill the ring with the subject rather than a lot of empty scenery, because the photo shrinks to a tiny size in feeds and chat lists. Strong, even light on the face helps it stay clear when the avatar is small. And before you settle, nudge the circle a touch in each direction to see which framing keeps the subject most balanced inside the round shape. --- # Compress Image Source: https://roundcut.app/compress/ > Drop one image or a whole batch, click Compress once, and download smaller files that keep their format. A JPG stays a JPG, a PNG stays a PNG. **Architecture:** hybrid, and the same path for one file or thirty. The grid holds your files locally; on Compress the bytes are POSTed to `/api/batch` with `format=keep`, a Cloudflare Worker forwards to a Fastify server running sharp 0.34 on libvips 8.17, each file is recompressed in its own format at a single server-chosen "recommended" quality, and the set comes back zipped through R2 behind a CDN URL. That stack produces materially smaller files than browser encoders — typically 10-15% better at the same visual quality, with tuned chroma subsampling browser APIs do not expose. The result is held in tenant-isolated storage for a short TTL and then deleted; the job id is what powers "delete now" and share. Uploaded files are never used for training, and the upload key is redacted from server logs. If the server lane throws, every file is recompressed in your browser instead, keeping its input format, and the zip is bundled client-side — the fallback is invisible, and the copy never claims "no upload". **Inputs:** JPG, PNG, WebP, AVIF on every modern browser. GIF (first frame) on Chrome/Firefox. HEIC on Safari only. Up to thirty files at a time. **Outputs:** the same format each file arrived in — a JPG stays a JPG, a PNG stays a PNG. One file on its own, or a zip for a batch. RoundCut Image Compressor makes your JPG, PNG, WebP, or AVIF smaller while keeping each file in its own format and at its original dimensions. Drop a single image or a whole batch onto the page, click the one Compress button, and download the result. You get back either a single smaller file or a zip of all of them together. Every file keeps the format you put in. A smaller JPG is still a JPG, so you can drop it straight back into the email, the listing, or the upload field it came from without anything breaking along the way. The width and height never change, only the size on disk goes down. When the run finishes the done screen shows you exactly how much smaller each picture got, measured from the real result, before you ever save it. ## What it does - Compress a JPG, PNG, WebP, or AVIF online - Compress one image or a whole batch at once - Every file keeps its original format - Up to thirty images in a single run - Dimensions stay the same, only the size drops - See exactly how much smaller your file got - A single image downloads on its own, a batch as one zip - Delete the working copy the moment you are done ## How to use it 1. **Add your images** — Drag one image or a whole batch onto the page, or tap the upload area to pick them from your device. 2. **Click Compress** — Press the single Compress button. There is nothing to set up, every file is handled at one tuned setting. 3. **Watch the progress** — A real progress bar fills as your images are worked through, so you always know how far along the run is. 4. **See how much you saved** — The done screen shows exactly how much smaller your file became, measured from the real result. 5. **Download the result** — Save a single smaller file on its own, or grab the whole batch as one zip, each file in its original format. ## Make any image smaller in one click This tool takes the images you already have and hands back smaller copies. Drop a JPG, PNG, WebP, or AVIF onto the page, click Compress, and download. There is one button and no settings to fiddle with. You are not stuck choosing a quality number or guessing what it means. Each file is recompressed at a tuned setting that keeps your picture looking right while cutting the size, and the done screen tells you how much smaller it ended up. When you need to change the width and height instead of the file size, run it through <a href="/resize/#tool">resize an image</a> first, then compress the result. ## Every file keeps its own format Compressing does not change what kind of file you have. A photo you put in comes back as the same type, just lighter, and a graphic keeps its own type too. That matters because the place you are sending your picture often expects one specific type, an email attachment, a shop listing, a website upload field. You get a smaller file you can drop straight back in without anything breaking. If you actually want a different type, that is a separate job, <a href="/convert/#tool">convert an image</a> changes one type into another, and you can compress the result afterward. ## Compress a whole batch at once You are not limited to one picture. Drop a stack of images onto the grid, up to thirty at a time, and the single Compress button handles all of them in one go. Each file is shrunk in its own format and handed back together as a zip, so there is no opening and saving one image after another. It is built for the moment you have a folder of photos for a listing, a gallery, or a set of email attachments. Drop a single image and you download that file on its own. When a picture also needs trimming, <a href="/crop-image/#tool">crop an image</a> first and then send the batch through. ## Smaller files, same dimensions Compressing lowers your file size without touching the width and height. A photo that is wide stays just as wide, it simply weighs less on disk and loads faster online. That keeps your picture sharp at the size it was meant for, which is what you want for a website, a document, or a print. When the dimensions are the real problem, a screenshot far bigger than it needs to be, the size drops much further if you <a href="/resize/#tool">shrink the dimensions</a> before compressing. The two steps stack neatly. Fewer dimensions first, then a lighter file. ## See exactly how much you saved When the run finishes, the done screen shows one clear number, how much smaller your file became. There is no math to do and no estimate, it is the real size of the file you are about to download. How big the drop is depends on the picture. A photo straight from a phone or camera usually shrinks a lot, while a file that was already squeezed once before has less room left and saves less. Now and then a heavily compressed source can come back about the same size, the picture was already near its floor. Either way you see the result before you save it. ## What it does and does not do It makes a smaller copy of your image while keeping the format and the dimensions, one picture or a batch, in a single click. The rest it leaves to other tools. It does not resize, so change the width or height with <a href="/upscale-2x/#tool">enlarge an image</a> or a downsize step when you need that. It does not change the format, that is a conversion. It does not edit your picture, so take out a background with <a href="/background-remover/#tool">remove a background</a> when that is the job. Camera details such as location and date are dropped when your file is rebuilt, which is usually welcome but worth knowing if you rely on them. ## Questions **Is the image compressor free?** Yes. You can compress as many of your images as you like, one at a time or a batch at once. Drop your files, click the Compress button, and download the result, there is no trial to start and nothing to install. The smaller file you get is yours to use however you need, on a site, in an email, or anywhere else you are sending it. **Will compressing change my file format?** No. Every file keeps the format you started with, the type you put in is the type you get back, just lighter. If you want a different type instead, that is a conversion, and you can swap the format with the convert tool and then compress whatever comes out of it. Compressing alone never touches the format itself. **Can I compress more than one image at a time?** Yes. Drop a stack onto the grid, up to thirty at once, and the single button shrinks all of them together in one pass. They come back zipped, each one in its own type. A lone picture downloads by itself rather than zipped, so it stays just as quick whether you bring one or a dozen along. **Which formats can I compress?** The four common photo and web types are all handled, and each one comes back smaller in the same format you put in, so the output always matches the input. Animated GIFs are not handled here, the convert tools cover those instead. A picture with transparency keeps its see-through areas after compressing, so a logo or a cutout still drops cleanly onto any background you place it on. **Does compressing make the image blurry?** Not in any way you would notice on a normal photo. The setting is tuned to cut the file size while keeping the picture looking the way it should at the size it is used. Where it can show is if you push the same file through again and again, each pass softens tiny details a little more. So compress once, starting from your original whenever you can, and you keep a clean picture at a much lighter size. **How much smaller will my file be?** A typical photo from a phone or camera often drops well past half its weight, and a graphic with flat colour can fall even further. The exact amount comes down to how much detail is packed in and whether anyone has trimmed it before. You will see the real figure for each picture you run, so there is never any need to estimate it up front. **Does compressing change the image dimensions?** No. The width and height of your image stay exactly the same, only the file size on disk goes down. When the dimensions are what you really need to change, a screenshot far bigger than it needs to be, resize the image first and then compress the smaller version. The two steps work together, smaller dimensions and then a lighter file, for the biggest drop of all. **What happens to my files after I compress them?** Your download is yours to keep, and the working copy used to build it is cleared shortly after the run, with a delete-now option if you want it gone the moment you finish. The compressed result keeps the picture itself but drops the camera details such as location and date, which is usually a good thing when you share a photo with anyone. ## The details ### Why a smaller file matters A heavy image is the quiet reason a page crawls, an email bounces back, or a chat refuses to send. Photos are by far the biggest thing most pages load, so a lighter file means a site that opens faster and a visitor who waits less, which search engines reward and people notice. Attachments have hard ceilings too, and a compressed photo slips under the limit a raw one blows past. The picture looks the same to the eye, it simply weighs a fraction of what it did, which is exactly what you want for the web, for mail, and for sharing. ### When a file barely shrinks Once in a while a file comes back close to the size you started with, and that is normal. It means the picture was already compressed hard before it reached you, a meme saved and re-saved, a screenshot a chat app squeezed, a download that was optimized on the way out. There is only so much room to remove, and a picture near that floor cannot give much more without the quality slipping. When that happens, the honest move is to keep the original, the gain is not worth a softer image. Start from the largest version you have for the best result. ### Built for one photo or a folder Some days you have a single picture to send, other days a whole shoot to get ready, and the same tool covers both. Drop one image and you get that file back on its own. Drop a folder and the one Compress button shrinks the lot in a single pass and hands them over as a zip, no saving one, then the next, then the next. It is the rhythm of real work, a product set for a listing, a gallery for a post, a run of attachments for an email. One click serves a single image and thirty alike. --- # Convert Image Source: https://roundcut.app/convert/ > PNG, JPG, WebP, AVIF. Pick a format pair below, drop your image, and get the converted file back. RoundCut's image converter changes a file's format across PNG, JPG, WebP, and AVIF. Pick a format pair from the grid below, drop a file onto the tool page, and the conversion starts immediately. Most single images convert on your device, so the result is ready in seconds. Converting to AVIF, or converting several images at once, runs on our server for the best result, and those files are processed and removed after about two hours. A batch download comes back as a single zip, with each file in the format you picked. PNG output keeps every detail exactly as it was, including any transparent areas. JPG, WebP, and AVIF compress the result, and at the default quality setting a typical photo shows no visible difference. If your image has a transparent background and you convert it to JPG, which cannot hold transparency, those areas become white. The converter accepts PNG, JPG, WebP, AVIF, and GIF as inputs, and for GIF files only the first frame is used. ## What it does - Converts between the four main web formats: PNG, JPG, WebP, and AVIF - A single image converts in your browser, on your device, with no upload - Batches of two or more are processed on the server with the link removed in about two hours - GIF accepted as input, with the first frame used for the conversion - Camera metadata stripped automatically during conversion - Original pixel dimensions preserved on every export - Real progress tracking when processing a batch on the server - Format grid updated as new pairs ship, with no copy changes needed ## How to use it 1. **Choose a format pair** — In the converter grid, tap the card that represents the conversion you want. The tool page opens with that format pair already selected. 2. **Add your image** — Drop the file onto the upload area or tap to choose from your device. PNG, JPG, WebP, AVIF, and GIF are accepted. For an animated GIF, only the first frame is used. 3. **Wait for the conversion** — A single image converts in the browser immediately, with no upload. A batch of two or more is sent to the server, which shows real progress as each file finishes. 4. **Check the result** — The done screen shows the converted image. If the target format was JPG and the original had transparency, those areas appear white. This is the expected behavior for a format that has no transparency channel. 5. **Download the converted file** — Tap download to save the file to your device. For a batch of several files, the download arrives as a ZIP. The download link for server-processed batches is removed after about two hours. ## How does the image converter work? Pick a format pair from the grid, then drop a file onto the tool page. The conversion starts the moment the file lands, with no Convert button to click. Most single images are decoded in your browser's memory, rewritten to the new format, and handed back as a download link. The original dimensions are preserved and camera metadata is stripped in the process. Converting to AVIF, or converting several files at once, sends the work to our server, which processes it and shows real progress as each file finishes. ## Which formats can I convert? RoundCut works with the four formats the web runs on. PNG keeps every detail without discarding anything and holds transparency, making it the right choice for logos and graphics with sharp edges. JPG compresses photos well and works everywhere, but has no transparency. WebP is the modern format that shrinks the file while keeping visual quality and preserving transparency. AVIF compresses further still at the same quality and has full support in current browsers. GIF works as an input, but only the first frame is used. The grid above shows every pair available today, and new pairs appear as they ship. ## Does converting affect image quality? It depends on the target format. PNG re-encodes without discarding anything, so every detail of the original is preserved. JPG, WebP, and AVIF compress when they save, and at the default quality setting the difference in a typical photo is hard to notice. Converting an already-compressed source back to an exact format like PNG does not recover what was discarded earlier. It just produces a larger file. If the source has a transparent background and the target is JPG, those areas become solid white because JPG has no transparency channel. ## Is it safe? Does my image get uploaded? Most single images convert entirely on your device, inside the browser. Converting to AVIF, or converting several images at once, sends the work to our server over an encrypted connection, processes it, and makes the result available to download for about two hours before the link is automatically removed. Those are the paths that upload your file, and you can clear the result yourself at any time. For the full detail on how files are handled, see the privacy policy. ## When should I change my image format? Switch format when something specific calls for it. Use WebP or AVIF when you want a smaller file for a website without sacrificing visual quality. Choose PNG when transparency must stay intact, as for logos and icons where every edge needs to be sharp. Pick JPG when broad compatibility matters more than modern compression, especially for older systems or content platforms that do not yet accept WebP. Avoid running the same image through several compressed formats in sequence, since each pass discards a little more and the result degrades. Convert once, to the format the destination actually expects. ## AVIF compresses the most, but takes longer to process AVIF produces the smallest files of the four formats: a typical photo can come out half the size of an equivalent JPG at the same visual quality. The trade-off is processing time. An in-browser AVIF encode can take several seconds on a desktop and longer on a phone, so converting to AVIF sends the file to our server, where it is generated much faster, and batches go there too. Other single conversions stay in the browser, which keeps AVIF fast on every device. ## Questions **Does converting reduce image quality?** PNG output keeps every detail exactly as it was. No information is discarded. JPG, WebP, and AVIF compress when they save, and at the default quality setting the difference in a normal photo is hard to spot. Converting a compressed source such as JPG into PNG does not recover lost quality; it just produces a larger file. **Is it safe to convert images online?** Most single images convert inside the browser, on your device. Converting to AVIF, or converting several at once, sends the work to our server, which processes it and removes the download link after about two hours. For how files are handled, see the privacy policy. **Can I convert multiple images at once?** Yes. Each format pair page accepts a single image or a batch, and the results come back together as a ZIP file. Most single images convert in your browser. Converting to AVIF, or a batch of two or more, is processed on our server and the download link is removed after about two hours. **What is the best image format for websites?** WebP and AVIF are the modern web formats, generally smaller than JPG at the same visual quality and supported in all current browsers. Use JPG for the widest compatibility across older systems and platforms. Use PNG when transparency needs to be preserved exactly, and AVIF when you want the smallest possible file for a modern audience. **What image formats can I convert?** RoundCut converts between PNG, JPG, WebP, and AVIF, the four main web formats. GIF works as an input using its first frame only, but is not available as an output yet. HEIC, TIFF, BMP, SVG, and RAW are not supported yet. New format pairs appear in the grid as they ship. **What happens to transparent areas when I convert to JPG?** JPG has no transparency channel. When you convert an image that has a transparent background to JPG, those transparent areas become solid white in the output. If you need to keep transparency, choose PNG, WebP, or AVIF as the output format, since all three support it. **Does converting remove camera metadata from the photo?** Yes. The conversion passes the image through a new encoding step, and camera metadata such as GPS location, device model, and exposure settings is stripped in the process. This happens whether the conversion runs in the browser or on the server. The color profile may be kept or removed depending on the browser and the output format chosen. **Why did my file get larger after converting?** PNG stores every detail without discarding anything. If the source was already compressed, such as a JPG or WebP, converting it to PNG re-encodes the result of that compression exactly, which produces a larger file. The larger size does not mean the output has more quality than the source had, only that the new format is more precise about what it received. ## The details ### How exact and compressed formats differ in practice PNG represents every point in the image with mathematical precision. When saved again, the output is bit-for-bit identical to the input. That is useful for logos, text screenshots, and anything that will be edited again, where each re-save must stay faithful. JPG, WebP, and AVIF work differently: when saving, they discard information the eye rarely notices, particularly in smooth gradients and uniform color areas. The file size reduction is real, but the discard is permanent. Re-wrapping a compressed file in an exact format recovers nothing and just packages the result in a larger file. This is why converting between compressed formats more than once accumulates visible degradation. The right approach is to save in an exact format while the image is still being worked on, and switch to a compressed format only at the end. ### Why WebP and AVIF are smaller without visible quality loss JPG was designed in the early 1990s for hard drives measured in megabytes. WebP and AVIF were developed decades later with more modern compression methods that better exploit how the eye perceives color and contrast. In practice, a WebP file at equivalent quality to a JPG is typically twenty-five to thirty-five percent smaller. AVIF goes further and for common photos can reach half the size of a JPG at visually indistinguishable quality. Both are supported in all current browsers, so choosing JPG today is mainly justified by compatibility with older systems or platforms that do not yet accept the newer formats, not by size or quality. ### Choose the format for the destination, not for the smallest size Reducing file size is a consequence, not a goal in itself. For an image going onto a website, WebP or AVIF are the natural choice because modern browsers understand them and pages load faster. For a file going to a designer or being edited further, PNG ensures nothing was discarded along the way. For print or for systems that only accept JPG, that format resolves the compatibility question with the widest reach. GIF as an input is useful when the only available source is an animated GIF and you need to work with the first frame. Converting with a clear intention, knowing where the image is going, avoids a second conversion that accumulates more discard and degrades the result. --- # Crop Image Hub Source: https://roundcut.app/crop/ > Circle, rectangle, or platform preset: pick the type that fits. Your image stays in your browser. Start from what the picture is for, and the right crop type follows. Need a round avatar or a contact photo? A circular crop hands back a clean circle with a transparent background, ready to save as PNG and drop into any round frame without white edges. Want to trim the edges, fix the framing, or hit an exact shape? A rectangular crop lets you drag a selection over the part worth keeping. When a platform demands precise dimensions, ratio locks such as 1:1, 16:9, and 9:16 hold the box proportional so the output never comes out stretched, and social-media presets go further by opening at the exact size the platform expects, so you skip looking up the numbers. The crop happens in your browser, on your device, with no upload step. Visit our /privacy/ page for full details on how your files are handled. Free to use. ## What it does - Round crop with transparent background for avatars - Free rectangular crop with manual framing - Fixed ratios 1:1, 16:9, 9:16, and more - Social-media presets with correct dimensions built in - Output in PNG, JPG, or WebP - Accepts PNG, JPG, WebP, AVIF, and GIF as input - Exports up to 4096 points per side - Crop runs in your browser with no upload step ## How to use it 1. **Choose the crop type** — On the page grid, select the shape you need: round, free rectangle, or fixed ratio for a social-media platform. Each option opens the corresponding tool. 2. **Open your photo** — Drag the image onto the tool or tap the upload area to select it from your device. PNG, JPG, WebP, AVIF, and GIF are all accepted. 3. **Frame the region** — Move and resize the crop box over the part of the image you want to keep. For a fixed ratio, the box moves while holding the locked proportion. 4. **Choose the output format** — Select PNG to preserve transparency (the recommended choice for circular crops) or JPG for a lighter file headed for the web. 5. **Download the result** — Click download and the cropped file is saved directly to your device, ready to use. ## Which crop type do you need? The shape you reach for depends on where the image is going. A round cut-out suits anything that shows up inside a circle (profile pictures, team avatars, contact thumbnails), and it leaves the corners transparent so the photo sits flat against any background without a white border. A free rectangle is the everyday choice for tightening a shot, removing a distraction at the edge, or re-centring the subject. When an upload form expects exact measurements, a fixed ratio or a named social-media size handles the math for you, keeping the proportions locked so the result lands correctly. Pick the one that matches the destination and the rest is a single drag. ## How cropping works in your browser The crop runs locally from start to finish. When you open a file, the browser opens and processes the image in your device's memory, you mark the area to keep, and that region is reconstructed and saved as a fresh file to download. The photo is cropped right there, immediately, with no upload queue and no wait for a server response. That keeps the flow fast and is especially practical for ID photos, screenshots, or any image you would rather handle directly on your device without passing it through a third-party service. The /privacy/ page explains how files are managed. ## Does cropping affect quality? Cropping does not blur or soften the image. You are keeping a section of the original at its native detail level and discarding the rest, so what stays looks exactly as it did before. The one thing to watch is final size: the crop is smaller than the source, so stretching a small crop to fill a large frame can look soft simply because there is less detail to cover that area. Crop close to the size you will actually display and the result stays sharp. Saving as PNG keeps every detail and any transparency. JPG trades a small amount of fidelity for a lighter file. ## What format should I save my crop? Match the format to the job. PNG is the right choice when the crop has transparent areas, a circular avatar above all, because it preserves the see-through corners the round shape leaves behind. It also keeps full detail, at the cost of a heavier file. JPG is the right call for rectangular photo crops headed for the web, where a smaller file loads faster and the slight trade in fidelity is invisible at normal viewing sizes. WebP sits in between, smaller than PNG and sharper than JPG at the same size, and supported across current browsers. When in doubt, PNG for round or detail-heavy crops, JPG for everyday photos. ## Crop vs resize vs compress These three get confused but they solve different problems. Cropping changes what the image shows by cutting away part of the frame. The composition changes, and what remains keeps its original detail. Resizing keeps the whole image and changes its dimensions, useful when something needs to be a specific width or height. Compressing keeps both the framing and the dimensions and only shrinks the file on disk by storing it more efficiently. Often you want more than one: crop first to fix the framing, then resize or compress for the final destination. Each lives in its own tool so you chain them in whatever order the task needs. ## Crop for social media and fixed ratios Every platform has its preferred dimensions, and uploading the wrong proportions means white bars, automatic trimming, or a stretched result. Fixed ratios like 1:1 for square profiles, 16:9 for covers and video thumbnails, and 9:16 for stories lock the selection to the right proportion while you adjust what appears inside. Social-media presets go a step further: the crop box opens at the exact dimensions each platform expects, so you skip the number lookup entirely. You frame what you want to show and download an image that uploads cleanly, without the platform auto-cropping your subject. ## Questions **Does cropping reduce image quality?** Not by itself. Cropping keeps a section of the original image at its native detail level. The only scenario where quality drops is if you then enlarge a small crop to fill a much larger frame. The image then has less detail to cover more area. Crop close to the size you plan to display and the result stays sharp. **Is it safe to crop images online?** Here it is, because nothing is uploaded. The crop runs inside your browser. The image is opened locally, never sent to a server, and cleared from memory when you close the tab. That makes it a good fit for private material such as ID documents, medical scans, or personal screenshots. See the /privacy/ page for full details. **Which crop shape should I use?** Let the destination decide. Use a circular crop for anything that displays in a round frame, such as profile pictures and avatars. It produces a transparent-background PNG without white edges. Use a free rectangle to retrim or recompose a photo. Use a fixed ratio or a social-media preset when the upload needs exact proportions so the result is not stretched or trimmed by the platform. **Can I crop a photo into a circle?** Yes. The circular crop produces a round image with a transparent background so the result sits inside any round frame without white corners. Save as PNG or WebP to keep the transparency. If you save as JPG the transparent corners fill with a solid color, since that format does not support transparency. **How do I crop an image for Instagram?** Use the social-media preset for the format you need: 1:1 for feed posts, 9:16 for stories and reels, or 4:5 for portrait posts. The crop box opens at the dimensions Instagram expects, so you adjust the frame until the subject is positioned the way you want, then download. No manual number entry needed. **Can I crop multiple images at once?** Each crop is one image at a time, because the framing is a manual choice that is rarely the same from photo to photo. Open one, position the crop, download it, then move to the next. There is no upload step between them, so working through a small set stays fast. **What image formats can I crop?** PNG, JPG, WebP, AVIF, and GIF. For animated GIFs, only the first frame is cropped. You can save the result as PNG when you need transparency or full detail, or as JPG for a lighter file. Pick the input you have and choose the output that fits where the crop is going. **Does the tool upload my files?** No. Your file is opened, cropped, and returned entirely on your own device, with no server call at any stage. Nothing is sent away, no copy is left online. You can confirm this in your browser's developer tools: the Network tab will show zero outbound requests while you crop. ## The details ### Why cropping does not blur your image When you crop a photo you are selecting a section of the original image and discarding the rest. None of that detail is altered, interpolated, or recreated during the process. What you see in the crop is exactly what the camera or app recorded in that region. Perceived softness only appears in one specific scenario: when the cropped area is small and you then enlarge it significantly to fill a large space, the existing detail has to be spread across a bigger area and the image can look less defined. The practical solution is to crop close to the final display size. If the image will appear 800 points wide, a crop at 900 points works well, but a crop at 200 points stretched to 800 will look soft. The crop operation itself degrades nothing. ### Choosing the right ratio for each platform Each platform handles image uploads differently, and uploading the wrong proportions results in automatic trimming that may cut out exactly what you wanted to show. The 1:1 square is the standard for profile pictures and feed posts on many platforms because it works equally well in wide and thumbnail views. The 16:9 ratio is the default for covers, banners, and video thumbnails where width dominates. 9:16 is the vertical format for stories and reels, designed for full-screen on a phone. Ratio locks fix the crop box to the correct relationship while you move and resize freely, without having to calculate width and height by hand. The result comes out in the format the platform expects, with no surprise trimming on upload. ### Crop, resize, and compress: when to use each The three operations change image files in distinct ways and often work together in sequence. Cropping removes part of the frame. The composition changes, but the detail that remains keeps exactly their original detail. Use it when the problem is what the image shows, not its size. Resizing changes the dimensions while keeping the whole image intact. Use it when an upload field needs a specific width or when the image is too large for its space. Compressing reduces the file on disk without touching the dimensions or framing. Use it when the image already has the right composition and size but is too heavy for email or slow to load. A typical sequence: crop, then resize, then compress. --- # Crop Image Source: https://roundcut.app/crop-image/ > Drag a selection to the exact frame you want, lock a ratio or a social preset, and download a clean cut as PNG, JPG, WebP, or AVIF. Crop an image down to exactly the frame you want, without uploading it anywhere. Drag a selection over your photo, lock it to a ratio like square, 4:5, or 16:9, or pick a social preset, and everything outside the box is trimmed away. The cut runs on your own device and the result downloads straight back to you, so there is nothing to wait on a server for. Export the crop as PNG, JPG, WebP, or AVIF, at the size of the part you kept. Choosing JPG fills any transparent areas with white, while PNG, WebP, and AVIF keep transparency. The selection is destructive, so the output is at most the source pixels you framed, never stretched bigger. For a round profile shape, use the circle crop tool instead. To learn how files are handled, see the privacy page. ## What it does - Crop any photo to a rectangle on your own device - Lock square, 4:5, 9:16, 16:9, 3:2, or 4:3 ratios - Start from Instagram, Facebook, YouTube, Twitter, or LinkedIn presets - Export as PNG, JPG, WebP, or AVIF - Nudge the selection a pixel at a time with the keyboard - Keeps transparency in PNG, WebP, and AVIF - Strips camera metadata from the cropped file - Crop one photo after another with no upload ## How to use it 1. **Open your photo** — Drag an image onto the page or tap to pick one from your device. It opens inside the crop frame right away, with nothing uploaded. 2. **Frame the crop** — Drag the handles and move the image to set the box over the part you want. Lock a ratio or a social preset if you need a set shape. 3. **Fine-tune with keys** — With the selection active, nudge it a tiny step at a time with the arrow keys to line an edge up exactly. 4. **Pick a format** — Choose PNG, JPG, WebP, or AVIF on the done screen. PNG, WebP, and AVIF keep transparency, JPG fills it with white. 5. **Download the cut** — Save the cropped image to your device. The file holds only the pixels you framed, with the camera metadata removed. ## How cropping works without uploading anything Pick a photo or drop it onto the page, and it opens inside the crop frame right away. Drag the handles to set the box over the part you want to keep, move the image underneath, and the area outside the selection is what gets trimmed. When the frame looks right, click crop and the cut downloads straight back to you. The whole cut runs on your own device, so nothing is sent to a server and there is no queue to wait in. You can crop one photo after another without the page ever reaching out to the network again. ## Lock a ratio for a profile picture or social post A freehand crop is handy, but most pictures need a set shape. Lock the selection to a square for a profile picture, to 4:5 for a feed post, to 9:16 for a story, or to 16:9 for a thumbnail, and the box keeps that ratio while you drag. Social presets go a step further and start the crop at the size a platform expects, like the square an Instagram post wants or the wide frame a video thumbnail needs. You frame the shot once and the result drops in at the right proportions, with no guessing and no second pass to fix the shape. ## Choosing JPG turns transparent areas white If your image has see-through parts, the output format decides what happens to them. PNG, WebP, and AVIF all carry transparency through, so the cropped corners stay clear. JPG has no transparency at all, so any see-through area is filled with solid white when you export to it. That is not a glitch, it is simply how the format works. Pick PNG or WebP when you need the transparency kept, and reach for JPG when the crop is a normal photo with no clear areas and you want the smaller file. The done screen lets you switch the format before you save. ## Nudge the selection with the keyboard Dragging with a mouse or finger is fine for a rough frame, but the last few steps are easier with the keyboard. Once the selection is active, the arrow keys move it a tiny step at a time, so you can line an edge up exactly without fighting the pointer. This also makes the tool usable when a mouse is hard to control, which matters for anyone who relies on the keyboard to get around a page. The crop box, the ratio lock, and the move all respond to keys, so a precise cut never depends on a steady hand. ## Export as PNG, JPG, WebP, or AVIF The crop can come out in four formats, and the right one depends on where the picture is going. PNG keeps every detail and transparency, which suits logos and graphics. JPG makes the smallest file for an ordinary photo, at the cost of transparency. WebP sits in between, smaller than PNG while still keeping clear areas. AVIF squeezes the file the most for the same look, though it takes a moment longer to write, especially on a cold start. Whatever you choose, the export holds the pixels you framed and nothing more, with the camera metadata stripped out along the way. ## When another crop tool fits better Cropping is about choosing a rectangle to keep. If you need a different shape or a different job, there is a closer tool. For a round profile photo with a transparent corner, use <a href="/#tool">circle crop</a> instead, since this tool only cuts rectangles. To set an exact width and height in pixels rather than a ratio, <a href="/resize/#tool">resize an image</a> is the fit. And if the file is simply too heavy after the cut, run it through <a href="/compress/#tool">compress an image</a> to bring the size down without touching the frame. ## Questions **How do I crop an image without losing quality?** Cropping itself does not soften what you keep, it only removes the area outside the selection, so the pixels inside stay exactly as they were. Quality drops only if you then export to a lossy format at a low setting, or expect the small crop to fill a much larger space. Keep the output as PNG, or as WebP at a high setting, and the part you framed looks the same as the original. **What aspect ratios and presets can I crop to?** You can crop freeform, or lock the selection to common ratios like square, 4:5, 9:16, 16:9, 3:2, and 4:3. On the social preset routes the crop also starts at the size a platform expects, such as a square Instagram post, a Facebook share image, a YouTube thumbnail, a Twitter header, or a LinkedIn banner. The box holds the ratio while you drag, so the frame never drifts. **Are my images uploaded to a server?** No. The crop runs entirely on your own device, and the photo is never sent anywhere. You pick a file, frame it, and the cut downloads straight back to you, with no round trip to a server in between. After the first load the page does not reach the network again, so a crop works even if the connection drops. The privacy page explains how files are handled. **Which image formats can I crop?** You can crop PNG, JPG, WebP, and AVIF images directly. Animated GIF is read as a single still, its first frame, since this is a still-image cropper. Formats like HEIC, TIFF, and RAW are not read here, so convert one of those to a common type first and then crop the result. The output comes back as PNG, JPG, WebP, or AVIF, whichever you choose. **Can I crop an image to custom dimensions?** Yes. You can drag a freeform selection to any shape, or lock a ratio and size the box to the part you want. The crop keeps the source pixels inside the selection, so the result is as large as the area you framed. It does not stretch a small selection up to a bigger target, which keeps the cut honest about how much detail it actually holds. **Can I rotate a photo while I crop it?** This tool focuses on the rectangular crop and the ratio lock rather than rotation, so it keeps the frame upright while you choose what to keep. If a photo comes in tilted, straighten or rotate it in an editor first, then crop the corrected version here. Keeping the two steps separate makes the crop itself fast and predictable. **Does cropping remove image metadata?** Yes. When the crop is exported, the camera metadata, such as the location, device model, and timestamps, is stripped out, and the file you download carries only the pixels. That is useful when you share a photo and do not want the embedded details going with it. If you need that metadata kept, work from a copy that still has it. **What is the difference between cropping and resizing?** Cropping changes which part of the image you keep by cutting away the edges, so the frame and the composition change but the kept pixels stay their original size. Resizing keeps the whole picture and changes its dimensions in pixels instead. If you want a tighter shot, crop. If you want the same shot at an exact width and height, <a href="/resize/#tool">resize</a> is the right tool. ## The details ### Why cropping on your device beats a server cropper A rectangular crop is light work for a browser, so there is no real reason to send the photo away to do it. When the cut runs on your own device, the picture never travels over the network, which means no upload wait, no queue behind other people, and nothing sitting on a server afterward. It also keeps working when the connection is patchy, because after the page loads the crop needs nothing more from the network. The trade is none, really, for this kind of edit: a server would only add a round trip and a copy of your file somewhere else, with no gain in the result. That is why the whole pipeline here stays in the page, from the moment you drop the photo to the moment the cut downloads. ### How a ratio lock saves a second pass Most pictures have a destination with a fixed shape: a square avatar, a 16:9 thumbnail, a 9:16 story, a banner that has to be wide and short. Cropping freehand and hoping it lines up usually means exporting, checking it in the destination, and coming back to fix the shape. A ratio lock removes that loop. You choose the shape before you frame, the selection holds it no matter how you drag, and the export lands at the proportions the destination wants on the first try. The social presets push this further by starting at the size in pixels a platform actually uses, so the crop is not just the right shape but the right size, ready to post without another tool in between. ### What a crop can and cannot recover It helps to know what a crop is doing to the pixels. Cutting away the edges loses nothing for the part you keep, because those pixels are passed straight through untouched. What a crop cannot do is invent detail: the result holds at most the source pixels inside your selection, so a tight crop of a small photo gives a small image, not a magically larger one. If you need the kept area to fill a bigger space, that is a separate job for an upscaler, and even then the extra pixels are guessed rather than recovered. Knowing this keeps expectations honest, and it is why the tool never offers to crop and enlarge in the same move. --- # JPG to PDF Source: https://roundcut.app/jpg-to-pdf/ > Drop in your photos, drag them into the order you want, pick a page size, and download a single PDF. Turn your photos into a single PDF without hunting for an app. Drop in one image or a whole set of JPG, PNG, and WebP files, drag them into the order you want, and they become the pages of one document. You can rotate any page a quarter turn, sort them by filename, and pick the page size, whether A4, Letter, or a page that matches each image. One image becomes a PDF right on your own device. Combining several sends them to a server that builds the document and keeps nothing afterward, so either way you only get the finished file back. Up to thirty images go into a single PDF. Each page is saved as a photo inside the document, so the result is meant for sharing and printing rather than editing text. To see how your files are handled, read the privacy page. ## What it does - Combine JPG, PNG, and WebP photos into one PDF - Up to thirty images in a single document - Drag the thumbnails to reorder the pages - Rotate any page a quarter turn - Sort every page by filename in one step - Choose A4, Letter, or an auto page size - Set the margin from none to wide - Camera metadata is left out of the finished file ## How to use it 1. **Add your photos** — Drag your images onto the page or tap to pick them from your device. Up to thirty go into one PDF. 2. **Set the order** — Drag the thumbnails into the sequence you want, and rotate any page a quarter turn if a shot is sideways. 3. **Choose the page** — Pick A4, Letter, or an auto size that matches each image, then set the margin from none to wide. 4. **Build the PDF** — Combine the photos into one document. A single image is handled on your device; several are assembled on a server that keeps nothing. 5. **Download the file** — Save the finished PDF to your device, ready to share, print, or send as one attachment. ## Combine many photos into one PDF Pick your photos or drop them onto the page, and each one becomes a page of a single PDF in the order you set. Drag the thumbnails to arrange them, then download the finished document. One image is turned into a PDF right on your device, with nothing sent anywhere. When you combine several, they go to a server that assembles the document and removes them once it is built, so the photos are not kept. Either way the result comes back to you as one file, ready to share or print, and you can start the next set straight after. ## Drag to reorder, rotate any page The order of the pages is yours to set. Each photo shows up as a card, and you drag the cards around until the sequence reads the way you want. If a shot came in sideways, rotate that page a quarter turn so it sits upright in the document. You can also sort every page by filename in one move, which is handy when your camera or scanner already numbered them. Nothing is locked in until you build the PDF, so you can rearrange right up to the last moment, and what you see in the grid is the order the pages land in. ## Pick the page size and margin A PDF has a page shape, and you choose it here. A4 and Letter give you the two standard sizes that printers and offices expect, while the auto option makes each page match the image on it, with no border at all. On top of that you set the margin, from none for an edge-to-edge look, to a small frame, to a wide one that leaves room to write or punch holes. The same choice applies across the whole document, so the pages stay consistent whether you are printing the result or sending it on. ## One honest note about image quality It helps to know what the tool does to each photo. Every page is saved as a photo inside the PDF at a high quality setting, not as a perfect copy of the original file. For sharing, printing, and sending, the difference is invisible, which is why this is the right trade for a document. What it is not built for is archiving a master image with every last detail intact, since the page is a saved photo rather than the source bytes. Transparent areas in a PNG are filled with white, because a PDF page has no see-through layer. ## JPG, PNG, and WebP all welcome You can mix the everyday photo formats in one document. JPG, PNG, and WebP all go in, and you do not have to convert them to match first, since each becomes a page the same way. A PNG with a transparent background lands on a white page, because the document format carries no transparency. Formats that are not photos, like HEIC from an iPhone or a TIFF scan, are not read here, so you can <a href="/convert/#tool">convert it</a> to a common type first and then add it. The camera metadata on each photo is left out of the finished PDF. ## Where your photos are handled It is worth being clear about where the work happens. A single image is turned into a PDF right on your own device, and that photo is not sent anywhere. When you combine two or more, they are uploaded to a server that builds the document and then removes the images, so they are not stored after the file comes back. Nothing is kept tied to you in either case. If you would rather shrink the result or change a photo first, run it through <a href="/compress/#tool">compress an image</a> or <a href="/resize/#tool">resize an image</a> beforehand. For the full detail, read the <a href="/privacy/">privacy page</a>. ## Questions **How do I combine JPG images into one PDF?** Drop your photos onto the page, drag the thumbnails into the order you want, and pick a page size. Then build the document and download it. One image is turned into a PDF on your device, while several are sent to a server that assembles them and keeps nothing. The result comes back as a single file either way, ready to share or print. **Is the JPG to PDF tool free?** Yes. You can build a document without signing up or installing anything, and there is no per-day cap on how often you use it. Drop in your photos, set the order and page size, and download the finished PDF. The file you get is yours to send, print, or upload wherever you need it. **How many images can I put in one PDF?** You can place up to thirty photos in a single document. Drag them into the order you want and each one becomes a page. If you have more than that, build a second PDF and combine the two in a dedicated PDF tool, since this one works from photos rather than merging existing documents. **Which image formats can I turn into a PDF?** JPG, PNG, and WebP all work, and you can mix them in the same document. A transparent PNG lands on a white page, because a PDF page has no see-through layer. Formats that are not photos, such as HEIC or TIFF, are not read here, so convert one of those to a common type first and then add it to the set. **Can I reorder and rotate the pages?** Yes. Each photo is a card you can drag into any position, so the pages end up in the sequence you choose. If a shot is sideways, rotate that page a quarter turn to stand it upright. You can also sort the whole set by filename in one step, which lines up pages that were already numbered by a camera or scanner. **What page sizes and margins are available?** You can set the document to A4, to Letter, or to an auto size where each page matches its image. Alongside that you choose the margin, from none for an edge-to-edge page, to a small frame, to a wide one. The setting applies across the whole PDF, so every page comes out consistent for printing or sending. **Are my images uploaded to a server?** It depends on the count. One image is built into a PDF on your own device, with nothing sent out. When you add two or more, they go to a server that puts the document together and removes them right after, so they are not stored once your file is back. You are never asked to sign in, and nothing is linked to you. The privacy page covers the specifics. **Will the PDF keep the full quality of my photos?** Each page is saved as a photo at a high quality setting, which looks the same as the original for sharing, printing, and sending. It is not a byte-for-byte copy, so this is not the tool for archiving a master image. Transparent parts of a PNG are filled with white, and the camera metadata is left out of the finished document. ## The details ### Why a PDF beats sending a folder of photos A handful of loose image files is awkward to send and easy to lose track of. They arrive in no fixed order, each opens on its own, and the person on the other end has to gather them back together. A single PDF solves all of that: the pages stay in the order you set, the whole thing travels as one attachment, and it opens the same way on a phone, a laptop, or a printer. That is why receipts, scanned forms, portfolios, and photo sets are so often asked for as one document. Building the PDF yourself also means you control the page order and the page size, instead of leaving it to whatever app the reader happens to use. ### What happens to a photo when it becomes a page When an image goes into the document, it is placed onto a page at the size and margin you chose and saved there as a photo. It is not pasted in as the original file, and it is not turned into selectable text, so the PDF is a faithful picture of each image rather than an editable copy. This is why a scanned letter made this way looks right but cannot be searched for words, and why a transparent PNG shows up on a white page. Knowing this sets the right expectation: the tool is built to package photos for sharing and printing, and a different kind of tool is needed when you want to edit the text inside a document. ### Why one image stays on your device and several do not There is a split worth understanding. Turning a single photo into a PDF is light enough to happen right in the page on your own device, so that image never has to be sent anywhere. Building a document from many photos is heavier, so when you add two or more they are uploaded to a server that assembles the pages and then removes the images once the file is ready. Nothing is kept tied to you, and the finished PDF comes back the same way in both cases. If a connection problem stops the upload, the page quietly falls back to building the document on your device instead, so you still get one PDF without having to retry. --- # Resize Image Source: https://roundcut.app/resize/ > Set an exact width and height or use a quick percentage, keep the proportions if you want, and download in the format you pick. Resize a photo to the size you actually need, without uploading it anywhere. Type an exact width and height, or tap a percentage like fifty or two hundred, and the image is redrawn at the new size right away. Keep the aspect ratio locked so nothing stretches, or turn it off when you need a precise shape, and switch on don't enlarge if smaller to stop a small picture from being blown up. Downscaling stays sharp and clean, which is what you want for a smaller copy. You choose the output as PNG, JPG, WebP, or AVIF on the way out, and the camera metadata is left behind. The whole resize runs on your own device, so the file is not sent to a server, and the result downloads straight back to you. For a tighter crop instead of a smaller copy, the crop tool is the better fit. To see how files are handled, read the privacy page. ## What it does - Resize to exact width and height on your own device - Quick percentages from a quarter up to double - Lock the aspect ratio so nothing stretches - Don't enlarge if smaller keeps small photos untouched - Sharp, clean downscaling - Save as PNG, JPG, WebP, or AVIF - Resize JPG, PNG, WebP, and AVIF inputs - Strips camera metadata from the saved file ## How to use it 1. **Open your image** — Drag a photo onto the page or tap to pick one from your device. It is ready to resize at once, with nothing uploaded. 2. **Set the size** — Type an exact width and height, or tap a percentage to scale down or up. Keep the aspect ratio locked so nothing stretches. 3. **Guard small photos** — Switch on don't enlarge if smaller when you want a small source left at its real size instead of stretched. 4. **Pick a format** — Choose PNG, JPG, WebP, or AVIF for the output. PNG, WebP, and AVIF keep transparency, JPG fills it with white. 5. **Download the result** — Save the resized image to your device. The file holds only the picture, with the camera metadata removed. ## Resizing a photo on your device, step by step Pick an image or drop it on the page, and it is ready to resize at once. Type the width and height you need, or tap a percentage to scale it down or up in one move. The new size is drawn right away so you can see the result before you save. The whole thing runs on your own device, so the photo is not sent to a server and there is no wait. When the size looks right, download it. You can resize one picture after another without the page reaching out to the network again after it loads. ## Hit the exact size a platform or avatar asks for Most resizing is about meeting a number someone else set. A profile photo wants a square of a certain size, a banner wants a specific width, an upload form rejects anything too large. Type the exact width and height and the image is redrawn to match, with the aspect ratio lock keeping it from stretching. When you only know one side, set that one and let the other follow on its own. The quick percentages are there for the common cases, so halving a photo for email or doubling a small graphic takes a single tap rather than mental arithmetic. ## PNG, JPG, WebP, or AVIF: which output to pick The format you save in decides the trade between size and fidelity. PNG keeps every detail and any transparency, which suits graphics and logos. JPG makes the smallest file for an ordinary photo, at the cost of transparency. WebP sits between the two, smaller than PNG while still holding clear areas. AVIF squeezes the file the most for the same look, though it takes a moment longer to write. You set the output on the way out, so you can resize once and save in whichever format the destination prefers, without starting over. ## Where resizing and compressing part ways Resizing and compressing both shrink a file, but they pull different levers. Resizing changes the dimensions: fewer pixels across and down, so the picture is physically smaller and lighter as a result. Compressing keeps the dimensions and trims the data instead, so the photo stays the same size on screen but weighs less. If a form wants a smaller width, resize. If the dimensions are fine but the file is too heavy to attach, <a href="/compress/#tool">compress an image</a> instead. Often the cleanest path is to resize first and then compress the smaller copy. ## What enlarging actually does to a picture Making an image larger is not the same as making it more detailed. When you scale up, the tool spreads the existing detail across more pixels and fills the gaps by blending neighbors, so the result stays smooth but no new detail appears. Past roughly double the size that blending starts to show as softness. The don't enlarge if smaller switch exists for exactly this reason, so a small source is left alone rather than stretched. When you genuinely need a small photo to fill a much larger space, an <a href="/upscale-2x/#tool">AI upscaler</a> is built for that job, since it invents plausible detail rather than just spreading what is there. ## Accepted formats and how your file is handled You can resize a JPG, PNG, WebP, or AVIF directly. Other types, like a HEIC photo from an iPhone, are not read here, so convert one of those to a common format first and then resize the result. Whatever you put in, the resize runs on your own device and the picture is not sent to a server. The result downloads straight back to you, and the camera metadata is left out of the saved file. If you would rather change the shape than the size, <a href="/crop-image/#tool">crop an image</a> is the tool for that. For the full detail on files, read the <a href="/privacy/">privacy page</a>. ## Questions **How do I resize an image without losing quality?** Making a photo smaller keeps it looking sharp, because the detail is packed into fewer pixels. To protect quality, leave the aspect ratio locked so nothing stretches, and save as PNG, or as WebP at a high setting, when you want every detail kept. Quality only drops noticeably when you enlarge a small photo well past its original size, which spreads the detail too thin. **Can I resize an image to exact dimensions?** Yes. Type the width and height you need and the picture is redrawn to match. With the aspect ratio locked, setting one side fills in the other on its own so the photo never stretches. Turn the lock off when a destination demands a precise shape that differs from the original, and the image is fit to those exact numbers. **What is the difference between resizing and compressing?** They sound alike but do different jobs. Resizing changes the dimensions in pixels, making the picture physically smaller. Compressing leaves the dimensions and trims the file weight instead. Pick resize when a form asks for a smaller width or height, and reach for <a href="/compress/#tool">compress</a> when the size is right but the file is too heavy to attach. **Which image formats can I resize?** You can resize JPG, PNG, WebP, and AVIF directly, and choose which of those four to save as on the way out. A HEIC photo from an iPhone is not read here, so convert it to a common format first and then resize the result. A transparent PNG keeps its clear areas when you save back to PNG, WebP, or AVIF. **How long does resizing take?** It is quick, because the work runs on your own device rather than on a server. For most photos the new size is drawn almost as fast as you set it. A very large source or an AVIF export takes a moment longer to write, but there is no upload to wait on and no queue, so the result is ready right after you choose the size. **Are my images uploaded to a server?** No. The resize runs entirely on your own device, and the photo is not sent anywhere. You pick a file, set the size, and the result downloads straight back to you, with no round trip in between. After the page loads it does not reach the network again, so a resize works even if the connection drops. The privacy page explains how files are handled. **Can I make an image larger?** You can, but enlarging spreads the existing detail across more pixels rather than adding any, so a much bigger copy looks softer. Past about double the size that softness shows. For a modest increase it is fine. When you need a small photo to fill a large space, an <a href="/upscale-2x/#tool">AI upscaler</a> is the better tool, since it builds plausible new detail. **Does resizing remove image metadata?** Yes. When the resized image is saved, the camera metadata, such as the location, device model, and timestamps, is left out, so the file you download carries only the picture. That is handy when you share a photo and do not want the embedded details going with it. If you need that metadata kept, work from a copy that still has it. ## The details ### Why downscaling looks sharp but upscaling goes soft Resizing down and resizing up are not mirror images of each other. When you shrink a photo, the tool has more detail than the new size needs, so it can average groups of pixels into clean ones and the result looks crisp. When you enlarge, the opposite is true: there are more pixels to fill than the source has detail for, so the tool blends neighboring values to bridge the gaps. That blending is smooth, but it cannot invent texture that was never captured, which is why an enlarged photo looks softer the further you push it. Knowing this, the rule of thumb is simple. Resize down freely, enlarge only a little, and reach for a different kind of tool when a small image truly has to fill a large frame. ### Aspect ratio, and why locking it matters Every photo has a ratio between its width and its height, and that ratio is what keeps faces and objects looking natural. When the aspect ratio is locked, changing one side updates the other to match, so the picture scales evenly and nothing is squashed or stretched. Unlock it only when a destination demands an exact shape that differs from the source, and be ready for the image to distort to fit. The safer habit for almost everything, from profile photos to product shots, is to leave the lock on and set just the side you care about. The don't enlarge if smaller option works alongside it, holding a small source at its real size instead of stretching it to reach a larger target. ### Resizing on the device, and what that buys you A resize is light work for a browser, so there is no reason to send the photo away to do it. Running on your own device means the picture never travels over the network: no upload wait, no queue, and nothing left on a server afterward. It also keeps working when the connection is patchy, because once the page has loaded the resize needs nothing more from the network. The trade is essentially none for this kind of edit, since a server would only add a round trip and a copy of your file elsewhere without changing the result. That is why the whole flow stays in the page, from the moment you drop the photo to the moment the new size downloads. --- # AI Image Upscaler Source: https://roundcut.app/upscale-2x/ > Make your photo sharper and bigger in seconds. The AI does the work in the cloud and hands back a result that is ready to download. Your photo comes back bigger and sharper in seconds. You drop in one image, and the AI rebuilds the edges and textures and hands back a higher-resolution version, ready to download. There is no fixed multiplier here. A small photo grows a lot, a photo that is already large mostly gets cleaner, and both land near the same size ceiling. The work happens in the cloud, so even a modest phone can handle it without installing anything heavy. Worth knowing: the AI does not bring back the original detail that was missing, it creates new detail that fits your image, which keeps the result looking natural. The photos that gain the most are the small, old, or low-resolution ones. If what you need is a lighter file instead of a bigger image, the compression tool is the right call. ## What it does - Raises image resolution with AI - Makes edges and textures sharper - Upscales one photo at a time - Accepts images in JPG, PNG, and WebP - Pulls photos from your device, Drive, or Dropbox - Rotate each photo before upscaling - Processes in the cloud in seconds - Downloads the upscaled photo ## How to use it 1. **Drop in your photo** — Drag an image onto the page or tap the upload area to pick from your device, your Drive, or your Dropbox. 2. **Check the grid** — See your photos as thumbnails, remove any you do not want, and rotate any that came in crooked before upscaling. 3. **Click upscale** — The AI sends your image to the cloud and starts raising the resolution. A bar shows the progress of each photo. 4. **Wait a few seconds** — The result appears as soon as the model finishes. Images upscaled before come back almost instantly. 5. **Download the result** — Save the upscaled photo to your device. With several photos, they come bundled in a zip, and you can clear the working copy from the same screen. ## How the AI upscaler works Your photo runs through an AI model trained on millions of images. It looks at the edges, textures, and colors of your image and generates new detail that blends with what was already there. The result comes back sharper and at a higher resolution, ready for you to download. Drop a photo onto the upload area, check it in the grid, and click upscale. Within a few seconds the result appears. If you like, you can rotate each photo before upscaling, or pull images straight from your Drive or your Dropbox. ## Output size: what to expect There is no fixed multiplier here. The AI always aims for a high final size, so how much your photo grows depends on where it starts. A small photo comes back much bigger. A photo that is already large comes back close to the same size, only with much firmer edges. Both cases land near the same resolution ceiling. That is why the tool never asks you to pick a scale factor by hand. It tunes the enlargement on its own to deliver the best possible quality for each image you send. ## Accepted formats and limits The tool accepts images in JPG, PNG, or WebP. You upscale one photo at a time. Each file has a ceiling on size and on input resolution, and photos above that are turned away before they upload, with nothing wasted. Phone-camera formats like HEIC, along with AVIF, GIF, and TIFF, are not accepted. If your photo is HEIC, you can <a href="/convert/#tool">convert the image</a> to a common format first and then upscale the result here. ## Your daily limit and what happens to your file The tool is open to use, with a daily cap on upscales per device to keep it standing for everyone. That cap is generous for everyday work, so you rarely bump into it during a normal photo session. If you do reach it, the screen tells you right away and asks you to come back the next day, without losing anything you already did. The counter runs per device, so a shared office network can add up faster than you would expect. And nothing you upscaled disappears when the day rolls over. ## Need a different result? Upscaling is for making a photo bigger and sharper. If what you need is something else, there is a right tool for each case. To change the width and height to an exact value, use <a href="/resize/#tool">resize an image</a>. To make the file lighter without touching the dimensions, run it through <a href="/compress/#tool">compress an image</a> after upscaling. Each one does a single job well, with nothing to install. The idea is always the same: one click does it, with no tangled menus or hidden steps along the way. ## Best photos to upscale Not every photo gains the same amount. The ones that gain most are the small, old, or low-resolution ones, like a cramped product shot, a portrait saved too tightly years ago, or a screenshot that came out blurry. In those cases the jump in sharpness is obvious. An image that is already large and well defined changes little, because there is almost nothing to improve. Before you upscale, it helps to start from the largest file you have of that photo, rather than a version that was already shrunk, so the AI has more to work with. ## Questions **What is an AI image upscaler?** It is a tool that makes a photo bigger and sharper using a trained AI model rather than plain stretching. You give it an image and it generates new detail that fits the photo, returning a version at a higher resolution. It runs in the browser interface and does the actual work in the cloud, so it takes only a few seconds even on a phone. You drop the image in and the result comes back ready. **Can I upscale an image without losing quality?** No upscaler enlarges an image with nothing changing, and this one is honest about that. The AI does not bring back the original detail that was missing, it generates new detail that matches your photo, making edges firmer and textures cleaner. The result looks bigger and sharper to the eye, even though it is not an exact copy of what the camera first captured. **Is AI image upscaling really free?** Yes, the tool is open to use and you pay nothing to upscale. There is a daily cap of photos per device, which keeps the tool standing for everyone. In everyday use you are unlikely to reach it. When you do, you simply come back the next day and keep upscaling whatever you still need to enlarge. **What image formats can be upscaled?** The everyday image formats of the web work directly. The usual trouble is an iPhone photo, which comes as HEIC and does not run here, along with a few less common types. The way around it is to convert your image to a common type before upscaling. Once that is done, it goes in like any other and the result comes out the same. **How long does AI upscaling take?** For a new photo, count a few seconds from the click to the result. If the same image has been through here before, it comes back almost instantly, because the service already has the finished result on hand. Bigger or more detailed photos ask for a touch more time, but the wait is short overall and you watch it on a progress bar. **Is my image safe when I upload it to an upscaler?** Not in the way you might fear. Once the upscale finishes, the result comes down to your browser and your file stops being tied to you or to any login. For a short while a cache version stays on the service, only so that re-sending the same photo does not have to run everything again. You can clear that result from the screen whenever you want. **What is the difference between upscaling and resizing?** Resizing changes an image's dimensions to a value you choose, and usually makes the photo smaller, inventing nothing. Upscaling with AI goes the other way: it enlarges the image and creates new detail so it can carry the larger size without going soft. If you just want an exact size, <a href="/resize/#tool">resize</a> handles it. To gain real resolution, upscale instead. **Does AI upscaling work on old or blurry photos?** It works best on photos that are small, soft, or low in resolution, which is exactly where old and lightly blurry shots fall. The AI firms up the edges and rebuilds some of the texture the blur ate away, so the image comes back more defined. If a photo is completely out of focus, though, there is not enough to work from and the gain stays small. ## The details ### Why upscaling with AI is not the same as stretching an image Stretching a photo in a plain editor just repeats the dots that are already there, and the result turns blocky and soft. Upscaling with AI is different. The model was trained on millions of pairs of small and large images, so it learned how a sharp edge or a patch of skin texture tends to look when it has more resolution. Instead of repeating dots, it invents new, plausible detail that fits what the photo already showed. That is why the result looks natural rather than blurred. Keep in mind that this detail is the model's educated guess, not an exact copy of what was in the scene. The image ends up bigger and cleaner to the eye, but it is not a perfect recovery of the original. ### When more resolution actually makes a difference Resolution becomes a problem the moment you print or upload to a place that demands a large image. A photo that looks great on a phone can come out jagged when you send it to print at a much bigger size, because there is not enough detail to fill the space. The same happens on marketplaces and catalogs that ask for product photos above a certain size. Upscaling before you send solves that squeeze without making you reshoot. On screen, the difference shows up when you crop a small piece out of a photo and still want that crop to look presentable. In all of these cases, what counts is not a scale number, but whether the final image can carry the size you need it at. ### Why the heavy lifting happens in the cloud An AI model that can redraw a whole photo is too heavy to run well on just any device. That is why your image goes up to a cloud service, which does the math on powerful hardware and returns the result in seconds, even if you are on a modest phone and without downloading anything large. That path comes at a cost: the photo has to leave your device to be upscaled, unlike a tool that only crops or resizes. In return, you get a result that would be slow or impossible to make in the palm of your hand. When it finishes, you can download the photo and clear the working copy from the same screen. --- # Convert AVIF to JPG, Open It in Any App Source: https://roundcut.app/avif-to-jpg/ > Turn an AVIF photo into a universal JPG and open it anywhere. Convert AVIF to JPG and open the photo anywhere. A single AVIF converts right on your device with no upload needed, and you get a JPG back in under a second for most photos since AVIF input decodes natively and fast. Converting several at once sends them to our server, which packs the results into one download, and whatever is sent is deleted within about 2 hours. AVIF is not accepted everywhere yet. Instagram uploads are inconsistent, Windows without the AVIF codec extension cannot preview the file, older Photoshop versions reject it, and most print services still want JPEG. JPG works everywhere. Transparent areas in the AVIF get filled with white in the output, because JPEG cannot store transparency, which is a property of the JPEG format, not a tool choice. The output JPG is usually larger than the source AVIF, since AVIF compresses more efficiently. That size bump is the cost of universal compatibility. ## What it does - Native AVIF decode. Your browser reads AVIF on its own with no module to download, so this side is fast and adds no page weight. - Native JPEG output. The browser's built-in JPEG writer handles the output, with no extra download for either step. - Single photo stays local. One AVIF converts on your device with no upload. Converting several at once uses our server, and what is sent is deleted within about 2 hours. - Under a second for most photos. A standard-size photo typically converts in under a second on desktop, using the fast native decode path. - Transparency filled with white. Clear or part-clear areas in your AVIF become white in the JPG, because JPEG cannot store transparency. - Opens anywhere. JPG works in every app, OS, social platform, email client, and print service with no codec to install. - Measured quality, 43.66 dB PSNR. The JPEG writer runs at near-maximum quality in our tests, visually identical to the source on photos. - Metadata dropped. The output JPG carries no EXIF, GPS, or camera fingerprint, only the visible image. ## How to use it 1. **Drop or pick your AVIF** — Click the upload area or drag an AVIF onto it. The conversion starts the moment the file lands, with no Convert button. A single file is handled on your device, while several at once go to our server. 2. **Let the conversion run** — For a single AVIF, the browser's native decoder and JPEG writer run on your device, and a standard-size photo usually finishes in under a second. Converting several at once runs on our server and returns one download. 3. **Read the output sizes** — A stats line shows the AVIF input size and the JPG output size side by side. Expect the JPG to be larger, since AVIF compresses more efficiently than JPEG. 4. **Download the JPG** — Click Download to save the result. The file keeps its original name and switches to the .jpg extension for you, so nothing needs renaming. ## How to convert AVIF to JPG Drag an AVIF onto the upload area, or click to pick one from your device. The conversion starts the instant the file lands, with no Convert button. A single AVIF is read by your browser's native decoder and written back out as JPEG right on your device, with nothing to upload. Drop several at once and they go to our server instead, which converts them together and hands back one download, with the files deleted within about 2 hours. When the JPG is ready, a stats line shows the input and output sizes, and the Download button drops it onto your device. For standard-size photos a single round trip usually takes under a second, since AVIF input rides the fast native decode path. ## What happens to transparent areas AVIF carries full transparency support, so its areas can be fully see-through or partly see-through. JPEG supports no transparency at all. Every transparent or part-transparent area in your AVIF has to be replaced with a solid colour before the JPEG is written, and the conversion fills those areas with white. If you need a different background colour. For example, if you want dark grey for a logo on a dark page, open the AVIF in an image editor, add a coloured background layer beneath the image, flatten the layers, and then convert the flattened file here. If you need the transparency itself to survive, convert to WebP or PNG instead of JPG. The white fill is a property of the JPEG format, not a setting this tool chose to hide from you. ## Why AVIF does not work everywhere yet AVIF is a newer format. Instagram and the wider Meta platforms have inconsistent upload support, with some workflows taking it while others silently reject it or re-compress it badly. Windows 10 and 11 need the AVIF Image Extension from the Microsoft Store to preview AVIF files in File Explorer and Photos, and without it the file shows up as a blank icon. Older Photoshop versions, before 2021, cannot open AVIF at all. Most stock agencies, print services, and e-commerce systems validate uploads against a MIME allowlist that predates AVIF. Email clients vary too, with Apple Mail showing it and Outlook on Windows not. Converting to JPG removes every one of these barriers in a single step, which is why compatibility, not quality, is usually the real reason to make this conversion. ## Why the JPG is usually larger than the AVIF AVIF and JPEG handle compression very differently. AVIF is a highly efficient modern format that shrinks photos dramatically. JPEG is the older, universally compatible format that trades compression efficiency for reach. So when you go from AVIF to JPEG, the output grows. In our measurements, AVIF files made from JPEG sources run 40 to 86 percent lighter than the original, and converting back brings the size close to what the original JPEG weighed. The JPEG writer runs at near-maximum quality, so the growth is entirely from format efficiency differences, not from re-compressing your photo at a lower setting. Think of it as paying bytes for compatibility: the trade is worth it when you need the file to open everywhere. ## Where your file is processed The answer depends on how many photos you bring. Drop one AVIF and the conversion happens entirely on your device: the browser's built-in image engine decodes the file and writes it as a JPEG, with nothing crossing the network. You can verify this yourself by checking the Network panel in your developer tools during a single conversion: the request log will show no outbound image traffic. Drop several AVIFs and they travel to our server instead, where they are converted together and returned as a single download. Whatever you send to our server is deleted within about 2 hours, and a clear-now button on the result screen lets you remove them immediately. Your photos are used only to produce your JPGs. Nothing is retained after the job is done. ## Does this tool handle animated AVIF? The converter keeps only the first frame of an animated AVIF. Every later animation frame is dropped, and the output is a still JPEG of that opening frame. JPEG itself has no animation support to fall back on. Animated AVIF sequences are usually short loops used in place of GIF for better quality and smaller files on the web. If you need to pull multiple frames out of an animated AVIF, a dedicated video or animation tool is the right choice. This converter's scope is single-image still conversion, and for a normal still photo none of this applies and the conversion behaves exactly as you expect. ## Questions **Why convert AVIF to JPG?** AVIF is not universally supported. Instagram uploads are inconsistent, Windows without the AVIF codec extension cannot preview the file in File Explorer, and older Photoshop versions and most print services reject it. JPG works on every device, every operating system, every social platform, and every editor. Converting trades AVIF's slimmer file size for guaranteed compatibility wherever you actually need to open or share the picture, which is usually the whole reason for the conversion. **What happens to transparent areas when I convert to JPG?** AVIF carries full transparency support, so areas can be see-through or partly see-through. JPEG cannot store transparency at all. The conversion fills every transparent area with white. If you need a specific background colour, paint it onto the AVIF in an editor before converting. If you need to keep the transparency itself, convert to WebP or PNG instead. Any tool that claims to preserve transparency when converting to JPG is incorrect, since the JPEG format makes that impossible. **What happens to my files? Are they uploaded anywhere?** It depends on how many you convert. A single AVIF is handled entirely on your device using the browser's own image engine, with nothing uploaded. You can confirm this by opening the Network panel in your developer tools during a single conversion: you will see no outbound image requests. Converting several at once sends them to our server, which converts them together and returns one download, and those files are deleted within about 2 hours, with a clear-now button on the result screen. We keep nothing once the job is finished. **What is the AVIF image format?** AVIF is a modern still-image format built on advanced video compression technology. It produces much smaller files than JPEG and WebP at the same visual quality, typically 40 to 86 percent lighter than JPEG. It was designed as an open, without licensing fees format. Browser support reached 94 percent globally in 2026, but many desktop apps, upload forms, and social platforms still do not accept it, which is why converting to JPG for compatibility is such a common task. **Is there a cost to use this converter?** Yes. Converting is available at no cost. You do not need to create an account or register. Your output arrives without any watermark. The only practical limit is your device's memory, which may restrict very large images on older hardware. A single image converts right on your device. Batches of several at once run on our server and results are deleted within about 2 hours. **Can I convert several AVIF files to JPG at once?** Yes. Drop several AVIF files and they are converted together on our server, then handed back as a single download. The files you send are deleted within about 2 hours, and you can clear them sooner from the result screen. A single file is instead handled right on your device with nothing uploaded. AVIF input decodes fast, so even a large batch finishes quickly. ## The details ### Why AVIF input is fast but AVIF output is slow One of the most common misreadings of AVIF is that it is always slow. Decoding AVIF is fast. The browser's native AVIF decoder, present in Chrome since version 85, Firefox since 93, and Safari since 16.4, runs in compiled native code with no JavaScript overhead. For a typical web-sized AVIF, native decode takes tens of milliseconds. Encoding AVIF is a completely different job. The codec's compression tests many prediction modes per block and runs expensive rate-distortion passes. When this site outputs AVIF, on the JPG to AVIF page, it uses a heavy software encoder. When it inputs AVIF, on this page, it uses the fast native decoder, with no module to download, no warm-up, and no several-hundred-millisecond init. That is exactly why AVIF to JPG conversion feels fast even though JPG to AVIF takes seconds. ### The transparency problem in detail: alpha, compositing, and white fill AVIF's transparency layer stores per-area opacity from 0, fully see-through, to 255, fully opaque. When the browser draws the AVIF onto a drawing surface and then encodes that surface as JPEG, the compositing step has to resolve each area's final colour. The surface starts as transparent black by default. The JPEG output path then fills the surface background with white before encoding, because JPEG must produce a fully opaque result. So every transparent region in your AVIF becomes white in the JPG. Partly transparent areas are composited onto white, so a region at 50 percent opacity lands halfway between its own colour and white. There is no setting to change this fill without drawing the compositing yourself onto a surface with a custom background, which is a job for an editor, not a converter. ### Quality measured: what 43.66 dB PSNR means in practice PSNR, Peak Signal-to-Noise Ratio, is measured in decibels and shows how closely a re-encoded image matches its source. Higher is better. In our end-to-end test run, the JPEG writer used here produced 43.66 dB PSNR on a standard photographic test image at the default quality. For reference, 36 dB is usually described as the threshold below which differences become clearly visible to most viewers under normal conditions, and 40 dB is generally rated perceptually transparent for photographic content. At 43.66 dB the output is visually identical to the source for photos viewed on screens at typical sizes. For graphics with sharp edges, fine text, or areas of extreme contrast, faint JPEG artifacts may show, since JPEG's block-based compression treats those regions differently from smooth photographic gradients. ### AVIF compatibility: where it works and where it still fails As of mid-2026, AVIF decodes in all major browsers, but compatibility with non-browser software is still patchy. On Windows, the AVIF Image Extension from the Microsoft Store adds support to Photos and File Explorer on Windows 10 and 11, and without it AVIF files appear as blank icons. Adobe added AVIF support in Lightroom 13.3 in 2024 and Photoshop 23.2 in 2022, so earlier versions cannot open it. Affinity Photo 2.3 from 2023 supports AVIF import and export. Paint.net needs the AVIF plugin at no cost. On macOS, Preview has read AVIF since Monterey. On phones, the system photo viewer supports AVIF on iOS 16 and Android 12 and later. The gaps are real and concentrated in enterprise tools, upload forms, and print workflows, which are exactly the situations where converting to JPG is the practical answer. ### AVIF versus HEIF: the difference and why it matters Both AVIF and HEIF, the High Efficiency Image File Format, store images using modern video codec compression. HEIF typically uses its own engine, while AVIF uses a newer, without licensing fees codec. The key distinction for web use is licensing. The codec behind HEIF carries per-device royalties that browser makers are reluctant to pay, which is why Safari supports HEIF but Chrome does not. The codec behind AVIF is without licensing fees by design, built by an industry alliance that includes Google, Mozilla, and Apple. That is why AVIF enjoys broader browser support than HEIF despite HEIF being the older format. iPhones have captured photos as HEIF since the iPhone 7, and those files carry a .heic extension. HEIC to JPG conversion is a different tool category from AVIF to JPG. This converter handles AVIF input only, so if you have a .heic file from an iPhone, you need the HEIC to JPG converter instead. ### When to keep AVIF and when to convert to JPG A useful rule: keep AVIF when you control both ends of the workflow, and convert to JPG when the file has to pass through a system you do not control. If you are delivering images on a site you built, with a CDN that supports format negotiation, AVIF is the better choice, being lighter, visually equal, and now covering 94 percent of browser traffic. If the image is heading into an email newsletter, a social upload, a form submission, a client-supplied print template, a shared folder others open on older software, or any legacy content system that validates file types, JPG is the safer choice. The practical pattern for most web projects is to store AVIF originals and produce JPG exports on demand for any situation that needs universal compatibility. --- # Convert AVIF to PNG, Universal Compatibility With Alpha Source: https://roundcut.app/avif-to-png/ > Convert an AVIF into a lossless PNG that opens in any app, with transparency intact. Convert an AVIF to PNG with transparency preserved and exact quality output. Drop one file and your browser decodes the AVIF natively using its built-in support, then re-encodes the result as PNG right on your device, with nothing uploaded. Convert several at once and RoundCut sends them to our server to do the work in one pass, and the download is removed within about 2 hours. There is no module to warm up in this direction, because both AVIF decode and PNG encode are native to every modern browser, so a single conversion finishes in well under a second for most images. Transparency is preserved throughout. Both AVIF and PNG support full transparency, so cut-outs, logos, and layered assets arrive in the PNG with the same transparent areas they had in the AVIF. The output is larger than the source by design, since PNG records every detail without lossy compression. Expect the PNG to run roughly three to ten times heavier. This pair helps when an AVIF refuses to open in an editor, email client, or platform that has not added AVIF support. ## What it does - Transparency preserved: every see-through area in your AVIF stays see-through in the PNG output. - Exact quality PNG output: the conversion adds no further quality loss beyond what the source AVIF already had. - Native decode with no warm-up: AVIF input is read directly by the browser, so this pair is fast on any engine. - One image converts right on your device with nothing uploaded, while converting several at once uses our server and the download is removed within about 2 hours. - Larger output by design: a PNG three to ten times the size of the source AVIF is expected, not an error. - Universal output: PNG opens in every viewer, editor, email client, and design tool, including pre-2025 Photoshop. - EXIF, IPTC and XMP are stripped, and ICC color profiles are kept in Chromium and WebKit but dropped in Firefox. - One AVIF in, one PNG out: single file per conversion in this version. ## How to use it 1. **Drop your AVIF** — Drag the AVIF onto the upload area, or click it to open a file picker and choose one from your computer, phone, or tablet. 2. **Wait for the result** — The browser decodes the AVIF natively and produces a PNG. No module needs to load, so the conversion is fast for most image sizes. 3. **Check the file sizes** — The stats line shows the original AVIF size and the new PNG size. Expect the PNG to run three to ten times larger than the source. 4. **Download the PNG** — Click the Download button to save the file with the same base name and a new .png extension, ready to open anywhere. ## How to convert AVIF to PNG Drop an AVIF onto the upload area or click it to select a file. The browser decodes the AVIF natively and re-encodes it as PNG using its built-in support. Because both steps are native, there is no module to load and no warm-up delay. For most photos the conversion finishes in under a second. When the result is ready, the stats line shows the original AVIF size next to the new PNG size, and the PNG will be noticeably larger. Click Download to save it to your device with the same base name and a new .png extension. The PNG is ready to open in any app on your machine immediately. ## Transparency is preserved Both AVIF and PNG support full transparency, so the conversion carries every transparent area through unchanged. Nothing is flattened and no background fill is added. If your AVIF holds a cut-out subject on a transparent canvas, the PNG you download has the same cut-out on the same transparent canvas, ready to layer into a design, a slide deck, or a web layout over any background. This is the opposite of converting to JPG, which has no transparency support and replaces every see-through area with a solid color. AVIF to PNG keeps the mask intact, which is the main reason to choose it over AVIF to JPG when your asset has transparency. ## Why is the PNG file larger? AVIF reaches its small size through compressed encoding, discarding information the eye does not notice and storing a compact representation. PNG keeps every detail exactly as decoded, with no further quality loss. When you convert AVIF to PNG, the converter expands the compact AVIF back into full image data and writes it without quality reduction, which is inherently larger. A typical ratio is three to ten times the AVIF size, depending on the content. This is not a sign anything went wrong. You trade file weight for universal compatibility and an exact intermediate that any tool can open and edit without another round of quality loss layered on top. ## When to use AVIF vs PNG Your choice comes down to who opens the file next. AVIF excels on a modern website where the audience uses up-to-date browsers and every saved byte improves page speed. PNG is the right pick when the destination does not yet read AVIF, including older viewers, email clients, Microsoft Word and similar office tools, forum software, and content platforms that reject AVIF on upload. PNG is also the right format when you plan to keep editing, because re-saving a PNG never adds further quality loss the way a re-saved compressed format would. AVIF is your delivery format, and PNG is your working and compatibility format. ## Where your file is processed It depends on how many files you convert. Convert a single AVIF and the whole thing runs on your device: the browser's built-in image engine decodes the file and writes it as PNG, with nothing crossing the network. You can verify this by opening the Network panel in your developer tools and converting one file: you will see zero outbound image requests during the operation. Convert two or more at once and RoundCut sends them to our server, which decodes and re-encodes them in one pass and returns a single download. Files handled on the server are processed and then removed within about 2 hours. We do not keep your images, we do not ask for an account, and we do not use them for anything other than the conversion you requested. Either way the output is the same PNG, with the same transparent areas intact. ## What does not yet support AVIF? AVIF is a recent format that many tools adopted only lately. As of 2026, these commonly reject it: most email clients including Gmail and Outlook, Microsoft Word and other Office apps, pre-2025 Adobe Photoshop, several content systems including some versions of WordPress and Squarespace upload widgets, some forum software, and every version of Internet Explorer. Modern browsers, Chrome 85 and later, Firefox 93 and later, Safari 16.4 and later, and Edge 121 and later, all read AVIF natively. If the tool you need sits in the first list, converting to PNG is the fastest way to make the file work where you need it. ## Questions **Why convert AVIF to PNG?** AVIF is not yet accepted by many tools. Email clients, older versions of Photoshop and other design apps, Microsoft Word, many upload fields, and forum platforms still reject it. PNG opens in every image viewer and app ever made. Converting to PNG is the fastest way to make your AVIF usable in a tool that does not support it yet. PNG is also the right choice when you need an exact working copy for further editing. **Does converting AVIF to PNG preserve transparency?** Yes, always. Both formats carry full transparency support, so the conversion is straightforward: every see-through area in your AVIF arrives in the PNG exactly as it was, with no background fill and no flattening. The key contrast is with AVIF to JPG, where transparency is impossible because JPEG cannot store it. Choosing AVIF to PNG instead is the right call whenever your image has a mask you need to keep, such as a cut-out product shot, a logo, or a layered design asset. **Will PNG files be larger than AVIF?** Yes, always, and by a wide margin. AVIF uses compressed encoding to shrink files dramatically. PNG records every detail without lossy compression. When you convert AVIF to PNG, the file grows by design, typically three to ten times the original AVIF size. This is expected behavior, not a fault. You are buying universal compatibility and an exact working copy in exchange for the larger file. **Do I lose any quality converting AVIF to PNG?** The conversion itself loses no quality. PNG records each area exactly as the browser decoded it from the AVIF. If the source AVIF was created with compressed encoding, that quality reduction happened when the AVIF was first encoded. Saving as PNG preserves what the AVIF contains but cannot recover information the original encoding already discarded. The PNG is a faithful copy of the decoded source, no better and no worse. **Why can't I open my AVIF file?** The format is newer than most software. Check if your app has a recent update: Photoshop added support in 2022, Lightroom in 2024, and Affinity Photo in version 2.3. If there is no update available, converting to PNG is the quickest path forward. PNG was standardised in 1996 and every imaging tool reads it without exception. On Windows, you can also install the AVIF Image Extension from the Microsoft Store at no cost, which adds native file support without converting. **Is it safe to convert my AVIF file here?** Yes. Converting a single AVIF produces a PNG entirely on your device with nothing uploaded, which you can confirm in the Network panel of your developer tools. When you convert a batch, files travel to our server, are processed together, and the results are removed within about 2 hours. RoundCut does not retain your images after the job is done, and no sign-in is required. If your images are sensitive, use the single-file mode and your data stays on your machine throughout. ## The details ### Why AVIF decode is native and fast Unlike AVIF encoding, which is heavy work, AVIF decoding is built directly into every modern browser. Chrome, Firefox, Safari, and Edge all ship a native AVIF decoder as part of their image pipeline, the same engine that renders AVIF embedded in web pages. So converting AVIF to PNG pays no warm-up cost and needs no extra module. The browser simply reads the AVIF, decodes it into an image buffer in memory, and hands that data to the native PNG encoder. The result is a conversion that finishes in well under a second for most real-world images, even large photographs. Only the PNG encode side varies in speed across browsers, and even the slowest observed path, around 122 milliseconds for a 1-megapixel image on WebKit, is fine for interactive use. ### The lossless paradox, why PNG is bigger It seems backwards that a conversion can make a file larger, but AVIF to PNG is exactly that case. AVIF reaches its small size by dropping visual information the eye does not notice at normal viewing distances, so its compression artifacts are real but invisible under ordinary conditions. PNG takes the opposite stance, storing every detail exactly as decoded with no further quality reduction, using only the non-lossy DEFLATE algorithm. The result faithfully represents the decoded AVIF image data without added degradation. Because the AVIF already discarded some detail, the PNG cannot be smaller than one made from the original source, but it is strictly true to what the AVIF held. Ratios of three to ten times the AVIF size are normal and expected, never a sign of error. ### Editing workflows that benefit from AVIF to PNG Designers and developers often receive AVIF files from automated image pipelines, content delivery networks, or modern content systems, then need to work with them in tools that have not added AVIF support. Converting to PNG yields an exact working copy carrying the same image data the AVIF held. From there, edits happen in Photoshop, Figma, Sketch, or any other design tool without adding compression artifacts. When the edited version is ready for the web, it can be re-exported as AVIF or WebP. This treats AVIF as the delivery format and PNG as the editing intermediate, the arrangement most compatible with current tooling. The PNG step adds no quality penalty beyond what was already present in the source AVIF. ### How alpha survives the conversion The transparency layer in AVIF is stored as a separate encoded plane alongside the color data. When the browser decodes an AVIF, it produces a color buffer and a transparency mask buffer. The conversion pipeline composites these at full transparency, then writes them out as a PNG. PNG's transparency encoding stores the mask as an 8-bit per-area channel attached to each color value. The result is a PNG where every area carries the same color and transparency values the browser read from the AVIF, ready for compositing. Soft transparency gradients survive. Feathered edges survive. Partially transparent areas survive. The only loss is whatever the AVIF's own compressed encoding introduced on the transparency plane when the AVIF was first created, usually a barely perceptible fringe at extreme zoom. ### When AVIF to PNG is right for web delivery Some web contexts genuinely require PNG even in 2026. Email newsletter images must be PNG or JPG because most email clients parse images on a remote server and reject AVIF. Social platforms vary: some process AVIF on upload and convert internally, others simply refuse the format. Print workflows running software PDF pipelines often need PNG for transparent layers, since AVIF is not part of the PDF imaging model. Open Graph preview images benefit from PNG or JPG because link-preview scrapers run on infrastructure that may lack AVIF decoding. For these delivery targets, converting to PNG is not a step backward but the correct choice given the destination system's constraints. The compatibility is worth the extra bytes. ### How processing differs from a typical online tool Most online AVIF converters upload every file you give them to machines you do not control, and hold the result under whatever retention policy the provider sets, often vague. RoundCut splits the work by how much you convert. A single AVIF is decoded and re-encoded entirely on your device, so nothing is uploaded for that case. When you convert two or more at once, the batch goes to our server, which does the decode and encode in one pass and returns a single download. Those files are processed and then removed within about 2 hours, without asking for an account and with no use beyond the conversion you asked for. There is no permanent storage and no sharing of your images. For developers handling client assets or proprietary imagery, the trade is straightforward: one image stays with you, a batch passes through our server only to produce your PNGs, and nothing lingers. --- # Convert AVIF to WebP, Broader Support With Alpha Kept Source: https://roundcut.app/avif-to-webp/ > Convert AVIF to WebP locally when you need broader app support without going all the way to PNG. Convert an AVIF to WebP fast. Converting one image happens locally on your device: it decodes the AVIF natively, then re-encodes it as WebP with the built-in encoder, so a single file is usually ready in under a second. Converting several at once is handled differently, as the batch is sent to our server to do the work and the download link is deleted within about 2 hours. Transparency is preserved either way, because both AVIF and WebP support a full transparent layer, so cut-outs and layered assets arrive in the WebP with the same masking they had in the AVIF. The output runs slightly larger than the source, typically 20 to 25 percent heavier, since AVIF compresses more efficiently than WebP at comparable quality. You convert for compatibility, not for size. WebP is accepted by email clients, older design tools, social upload pipelines, and content systems that do not yet handle AVIF. ## What it does - Transparency preserved, every transparent area in your AVIF arrives intact in the WebP, with no flattening or solid-color fill. - Fast native conversion, AVIF decode and WebP encode are both native, with no module to load and no warm-up delay. - Convert for compatibility, WebP is accepted by platforms and tools that still reject AVIF, the output is slightly larger by design. - A single image converts locally on your device with nothing sent anywhere, while converting several at once is handled on our server and the download is deleted within about 2 hours. - WebP slightly larger than AVIF, expect roughly 20 to 25 percent heavier, the trade for broader support. - One re-encode step, near-lossless settings measure around 44 dB PSNR on photographs, visually identical at normal sizes. - EXIF, IPTC and XMP are stripped, while ICC color profiles hold in Chromium and WebKit but drop in Firefox. - One AVIF in, one WebP out, single file per conversion in this version. ## How to use it 1. **Drop your AVIF** — Drag the AVIF onto the upload area, or click it to open a file picker and choose one from your device to begin. 2. **Wait for the result** — The browser decodes the AVIF natively and produces a WebP. Both steps are native, so no module loads and the conversion is fast. 3. **Review the sizes** — The stats line shows the AVIF source size and the WebP output size. Expect the WebP to run roughly 20 to 25 percent larger than the AVIF. 4. **Download the WebP** — Click the Download button to save the file to your device with the same base name and a new .webp extension. ## How to convert AVIF to WebP Drop an AVIF onto the upload area or click it to select a file. The browser decodes the AVIF natively and re-encodes the content as WebP with its built-in encoder. Both steps are native, so there is no module to load and no warm-up delay. For most photos on a desktop browser, the conversion finishes in under a second. When the WebP is ready, the stats line shows the original AVIF size next to the new WebP size. Click Download to save it to your device with the same base name and a new .webp extension. The WebP is ready to hand off to any platform that accepts the format immediately. ## Why is the WebP file larger than the AVIF? AVIF is the most efficient mainstream image format in 2026, typically reaching 20 to 25 percent smaller files than WebP at the same visual quality. Converting from AVIF to WebP moves from a more efficient format to a less efficient one, so the output is inherently larger. This is expected and is not a sign anything went wrong with the conversion. You make a deliberate trade: giving up a little file efficiency in exchange for WebP's broader acceptance across platforms and tools. The output WebP is still far smaller than a PNG of the same image, so your file stays compact. The cost is modest relative to the gain in compatibility across the systems that have not caught up with AVIF yet. ## When to use AVIF vs WebP Keep AVIF when you control delivery end to end, your audience uses modern browsers, and every kilobyte counts. Convert to WebP when the destination does not yet accept AVIF: WordPress and Shopify media libraries with outdated upload handlers, social platforms that process images on remote servers with older codecs, email newsletter platforms where images are pre-processed, legacy delivery configurations, older design tools, and any platform where you have hit AVIF rejection. You are not downgrading quality in a meaningful way, since WebP at high-quality settings looks the same as the AVIF at normal display sizes. You are buying access to the systems that have not yet caught up with AVIF support. ## Does transparency survive AVIF to WebP? Yes. WebP supports transparency, and so does AVIF, so the conversion keeps every transparent area intact. Logos with soft drop shadows, product cut-outs with feathered edges, UI assets with rounded corners, all arrive in the WebP with the same masking they had in the AVIF. This is the key difference from converting to JPG, which has no transparency and replaces it with a solid fill. Both the color data and the transparency mask are re-encoded at high-quality settings, so the edges stay clean. No manual flattening of your image is needed before the conversion. ## WebP browser and platform support WebP is supported by essentially every modern browser: Chrome, Firefox, Safari, Edge, and all major mobile browsers. Global support sits above 97 percent. More important for this use case, WebP is accepted by a much wider range of platforms than AVIF. Email clients that reject AVIF often handle WebP correctly when the server pre-processes it. Content systems that block AVIF uploads typically accept WebP. Design tools that have not added AVIF read WebP. If your workflow takes images through multiple systems before delivery, WebP is the safer intermediate today. For pure browser-to-browser delivery AVIF is more efficient, but the compatibility story for WebP across non-browser systems is genuinely stronger. ## Where your file is processed It depends on how many files you convert. Converting one image runs entirely using your device's native image engine: the AVIF is decoded and the WebP is encoded on the spot, with nothing sent anywhere. Converting several at once works differently: the batch goes to our server, which does the encoding and zips the results, and the download link is deleted within about 2 hours. We hold no permanent copy of your image beyond that short window. So a single conversion is local from start to finish, and a batch is processed on our server and then cleared, with nothing of yours retained afterward. ## Questions **Why convert AVIF to WebP?** WebP is accepted by a wider range of systems than AVIF. Email clients, older upload handlers like some versions of WordPress and Shopify, social platforms that process images on remote servers, legacy delivery configurations, and design tools that have not added AVIF support all handle WebP. You already have AVIF for best compression. You need WebP when the platform receiving your image does not read AVIF yet. The reason is compatibility, not file size, and the cost is a slightly heavier file. **Is WebP smaller than AVIF?** No. AVIF is typically 20 to 25 percent smaller than WebP at the same visual quality. Converting AVIF to WebP produces a slightly larger file, not a smaller one. If minimizing file size is the goal and the destination supports AVIF, keep your AVIF. Convert to WebP only when the destination requires it. The WebP output is still far smaller than an equivalent PNG, so it remains a compact intermediate for compatibility. **Does converting AVIF to WebP preserve transparency?** Yes. Both AVIF and WebP support transparency, so the conversion keeps every transparent area intact. Nothing is flattened and no background color is added. This differs from converting to JPG, which has no transparency support and would flatten every transparent area to a solid color. Your logos, cut-outs, and UI assets keep their soft edges and rounded corners after the conversion, ready to layer over any background you place behind them. **Will converting AVIF to WebP reduce quality?** Very little. The WebP encoder runs at a high-quality setting that measures around 44 dB PSNR on photographs, visually indistinguishable from the AVIF source at normal display sizes. There is one re-encode step, so the result does discard a small amount of detail, but artifacts are not visible to the naked eye on natural images. For graphics with very sharp edges or hard color transitions, check the output carefully since those areas are more sensitive to any compression change. **How long does AVIF to WebP conversion take?** For most photos on a desktop browser, the conversion finishes in under a second. AVIF decode and WebP encode are both native to modern browsers, with no module to load. A typical 2-megapixel photo finishes in about 100 to 200 milliseconds on Chrome. Even large 4K photographs usually complete in under a second. This is much faster than the reverse direction, AVIF encoding from a WebP, which has to load a heavy encoder and do far more intensive work. **Which browsers support WebP?** WebP is supported by essentially all modern browsers and has been for years. Chrome has supported it since version 23, Firefox since 65, Safari since 14, and Edge since 18. Mobile browsers follow the same versions. Global WebP support passes 97 percent. In practice, if a browser is in active use in 2026, it almost certainly reads WebP. The remaining edge cases are Internet Explorer and very old versions of Safari on iOS 13 or earlier, which are rare today. ## The details ### The compatibility landscape for AVIF in 2026 AVIF browser support reached about 94.3 percent of global browsers by 2026, but browser support is not the full picture. A large share of image consumption happens outside browsers: email clients rendering inline images, design tools opening files for editing, content systems validating and re-processing uploads, delivery image pipelines, document editors embedding assets, and social platforms that process images at upload time. In most of these non-browser systems, AVIF support lags well behind browser adoption. Gmail, Outlook, and most corporate email clients still process images through older pipelines that reject AVIF. Adobe Creative Cloud added AVIF support only in late 2024 versions. Many WordPress configurations with older image plugins still block AVIF on upload. WebP, by contrast, has been adopted across nearly all of these platforms for years. Converting AVIF to WebP is the compatibility bridge for this non-browser infrastructure. ### Why the conversion is fast on both sides The speed advantage of AVIF to WebP over the reverse direction comes from the codec architecture. Decoding AVIF is handled by a native browser decoder that runs with hardware acceleration on modern devices. Encoding WebP uses the browser's native WebP encoder, also hardware-accelerated on most platforms. Neither operation needs to load a heavy module, which is the bottleneck for AVIF encoding. The encoder for AVIF output is large and needs about a second to initialize per session. AVIF to WebP skips all of that. The whole round trip for a 2-megapixel photo finishes in well under a second on any modern desktop or laptop browser. This makes AVIF to WebP suitable for interactive workflows where the user expects a response in under a second. ### How much the re-encode actually costs Converting AVIF to WebP involves one re-encode step. The AVIF was originally encoded with some degree of compression. Decoding it gives color values that reflect that source. The WebP encoder then applies its own compression to those values at a high-quality setting tuned to quality 85. At that setting, the output measures around 44 dB PSNR on typical photo content. For a viewer looking at a photograph at normal display size, the difference between the AVIF source and the WebP output is not visible. For graphics with very fine text at small sizes, precise icons, or hard-edged color blocks, the cumulative effect of two compression passes can show subtle differences under close inspection. Before committing to a full library conversion, test a representative sample at full zoom on your most quality-sensitive assets. ### The transparency round-trip in detail The transparency layer in AVIF is stored as a separate encoded plane. When the browser decodes an AVIF, it produces both a color buffer and a transparency mask. The conversion composites both at full transparency, preserving every partially transparent area. The WebP encoder then writes a file with a separate transparency layer encoded using WebP's own algorithm for that plane. The result is that the transparency mask in the output WebP is stored relative to the decoded values from the AVIF. Soft gradients and feathered edges survive. The only change present is what the AVIF's original encoding introduced. If the source AVIF has clean transparent edges, the WebP output will too, with the same masking ready for compositing over any background. ### Comparing the output to the alternatives When you need to make an AVIF compatible with a system that does not read AVIF, you have three realistic options: convert to WebP, convert to PNG, or convert to JPG. JPG is the wrong choice for any asset with transparency, because JPG has no transparency support and flattens it to a solid color. PNG produces the largest file, typically three to ten times the AVIF size, and is the right choice only when you need an exact intermediate or the destination requires PNG specifically. WebP sits in the middle: it gives universal modern compatibility, keeps transparency, and produces a file usually 20 to 25 percent larger than the AVIF rather than 300 to 1000 percent larger like PNG. For any compatibility conversion that does not require an exact output, WebP is the right intermediate. ### Single in the browser, batches on a server This pair runs two ways depending on the job. A single AVIF is decoded and re-encoded as WebP entirely inside your device using native paths, so for one file there is no upload at all. That is the right path for a quick one-off and for confidential client work, proprietary product images, or documents you would rather keep on your own machine. Converting several files at once is handled on our server instead, because batching, zipping, and delivering a set is the job a server does well: the files are sent up, encoded, packaged, and handed back as one download, which is then deleted within about 2 hours with no long-term storage. The practical read is that a single conversion is local, and a batch is processed remotely but kept only for the brief window it takes to download. --- # Convert JPG to AVIF, the Smallest Modern Photo Format Source: https://roundcut.app/jpg-to-avif/ > Turn a JPEG into a featherweight AVIF and shrink photos by 40 to 86 percent. Convert JPG to AVIF and the encoder produces a file that usually runs 40 to 86 percent lighter than the source at the same visual quality. In our tests a small JPEG became 65 percent smaller as an AVIF, and a large photo dropped by 86 percent. To get the best AVIF, RoundCut may process the image on our server, and converting several at once always runs there. Whatever is sent is deleted within about 2 hours, and you can clear it sooner. If the server path is unavailable the conversion finishes on your device instead. AVIF is a compressed format like JPEG, so this is a compressed-to-compressed pass, and at the default quality the output is indistinguishable from the source on photographs. Large photos above 8 megapixels can take several seconds to encode, since AVIF trades encode speed for better compression. ## What it does - Tuned AVIF encoder. Produces a file 40 to 86 percent lighter than JPEG at indistinguishable quality. - 40 to 86 percent lighter than JPEG. Our tests show small JPEGs shrink by 65 percent and large photos by 86 percent. - Short-lived processing. To build the best AVIF we may handle the file on our server, and converting several always does, with whatever is sent deleted within about 2 hours. - Default quality tuned for photos. Around 42 dB PSNR, visually identical to the source at normal viewing distances. - Warms up once per session. After the first encode the encoder stays cached, so later conversions in the tab start right away. - 94 percent browser support. Chrome 85+, Firefox 93+, Safari 16.4+, and Edge 121+ all read AVIF natively in 2026. - Metadata dropped. The AVIF output carries no EXIF, GPS, or camera fingerprint, only the visible image. - Auto-renamed download. The file keeps its original name and gains the .avif extension for you. ## How to use it 1. **Drop or pick your JPEG** — Click the upload area or drag a JPEG onto it from your device. The conversion starts on its own the moment the file lands, with no separate Convert button. 2. **Let it process** — To build the best AVIF, RoundCut may run the encode on our server, and converting several always does. A single photo may instead finish on your device, taking a moment to warm up the first time. 3. **Let the encode finish** — Small photos finish in a fraction of a second. Photos above 4 megapixels take a few seconds on desktop, and mobile runs slower. A progress indicator shows during the encode. 4. **Download the AVIF** — Click Download to save the result. The file keeps its original name and switches to the .avif extension for you, so nothing needs renaming. ## How to convert JPG to AVIF Drag a JPEG onto the upload area, or click to pick one from your device. The conversion fires the moment the file lands, with no Convert button. To produce the best AVIF, RoundCut may send the image to our server, and converting several at once always does, with the result deleted within about 2 hours. If that path is unavailable the encode finishes on your device instead, which can take a moment to warm up the first time. Small photos finish in a fraction of a second. A photo at 8 megapixels takes around 2 to 3 seconds. When the AVIF is ready, click Download to save it with the same base name and the new extension. ## How much smaller will the AVIF be? In our end-to-end tests, a typical small JPEG became 64 percent smaller as an AVIF, and a large photo at high resolution became 86 percent smaller. The realistic range on real photographs is 40 to 86 percent lighter, depending on what is in the frame. Smooth areas like skies and skin compress hard. Busy textures like foliage, fabric, and film grain compress less. The quality sits at the default tuned for photography, with PSNR around 42 dB, which is visually identical to the source at normal viewing distances. AVIF's lead over WebP widens at lower quality settings and on photographic content carrying large, smooth regions where the prediction model really pays off. ## Does AVIF work everywhere? AVIF reads in 94 percent of browsers worldwide as of 2026. Chrome since version 85, Firefox since 93, Edge since 121, and Safari on iOS 16 and macOS Ventura and later all decode it natively with no plugin. Older iPhones on iOS 15 or earlier do not, and neither do Internet Explorer or pre-2022 Android browsers. For full coverage, serve AVIF with a JPG or WebP fallback inside the HTML picture element, and the browser picks the first format it can read. For modern web delivery where your audience is on recent devices, AVIF works on its own with no fallback needed. The format is past the experimental stage and squarely into production-ready territory for general consumer traffic. ## When to use AVIF versus JPG Reach for AVIF on hero images, gallery photos, and anything served from a CDN that supports content negotiation, where your audience is on modern browsers. AVIF wins on file size at any quality compared with both JPEG and WebP. Keep JPEG for legacy pipelines: email attachments, CMS uploads that downstream tools may re-process, print services, and any workflow built before 2022. Most production sites today serve AVIF with a JPEG fallback through the picture tag, banking the size win for modern browsers while keeping full coverage for the rest. If you are replacing JPEG images on a site you control, AVIF is the strongest single-format choice in 2026, provided you can absorb a slower one-time encode. ## Where your photo is processed AVIF is the slow, heavy format, so to give you the smallest file at the best quality RoundCut may encode it on our server rather than on your device. Converting several at once always runs on our server, where the files are zipped and handed back as a single download. Anything sent there is deleted within about 2 hours, and you can clear it right away from the result screen. When the server path is unavailable, the same conversion runs on your device as a fallback. So a single photo may be processed locally or may go to our server depending on conditions, and converting a batch always uses it. Either way the photo is used only to make your AVIF and nothing is kept. ## Is AVIF better than WebP? AVIF and WebP are both modern, without licensing fees formats that compress better than JPEG. AVIF generally hits smaller files at the same quality, especially on smooth photographic content, and it supports HDR and 10-bit colour depth. The trade-off is encode speed. AVIF is markedly slower to encode than WebP. Turning a 4K photo into AVIF takes around 2.8 seconds on desktop Chrome, against under 600 milliseconds for WebP. Browser support is close, with WebP at 97 percent of traffic and AVIF at 94 percent. For web delivery where you want the lightest possible files and can absorb a longer one-time encode, AVIF is the better pick. For speed-critical batch work, WebP stays the more practical choice. ## Questions **Why convert JPG to AVIF?** AVIF produces the smallest files of any widely supported format at a given quality, typically 40 to 86 percent lighter than JPEG and smaller than WebP at the same setting. If you manage hero images or gallery photos on a site, moving to AVIF can trim your bandwidth costs, lower hosting expenses, and directly improve Largest Contentful Paint, a Core Web Vitals metric Google uses as a ranking signal. Browser support now sits at 94 percent globally, so most visitors will see the AVIF version. **What browsers support AVIF?** Chrome 85 and later, Firefox 93 and later, Safari 16.4 on iOS 16 and macOS Ventura and later, and Edge 121 and later all decode AVIF natively. That covers roughly 94 percent of browsers worldwide in 2026. The remaining 6 percent is mostly older iOS devices on iOS 15 or earlier, Internet Explorer, and some dated Android browsers. For full coverage, pair the AVIF with a JPEG or WebP fallback inside the HTML picture element and let each browser choose. **Is AVIF better than WebP?** For file size, AVIF generally wins, since it hits smaller files at the same visual quality, especially on smooth photographic content. For encode speed, WebP is much faster, encoding in under 600 milliseconds for a 4K photo while AVIF takes around 2.8 seconds on desktop Chrome. Browser support is close but WebP leads slightly at 97 percent versus 94 percent. For production web delivery where size matters most, AVIF is the better choice in 2026. For real-time or batch encoding where speed matters, WebP is more practical. **Does AVIF support transparency?** Yes. AVIF supports transparency, which JPEG does not. When you convert from a JPEG source, though, there is no transparency to keep, since JPEG is always fully opaque. If you start from a PNG or WebP that has transparent areas and want them preserved in AVIF, convert from that source directly instead. This tool's AVIF output will carry the transparency information whenever the source image already has it. **What quality setting does it use?** This tool encodes at a near-maximum quality default tuned for photography. The resulting PSNR sits around 42 dB, which is visually identical to the source for photos under normal viewing conditions. That setting balances file size and quality well: going lower would shrink files further but introduce visible artifacts on photos, while going higher would erode the size advantage that makes AVIF worth using. The default we ship is the right setting for most web publishing, which is exactly why there is no slider to fuss with. **Can I convert several JPGs to AVIF at once?** Yes. Drop several JPEGs and they are converted together on our server, then handed back as a single download. The files you send are deleted within about 2 hours, and you can clear them sooner from the result screen. A single photo may instead be handled on your device. AVIF encoding is compute-heavy, so a large batch takes a little time to finish. ## The details ### Why AVIF encode is slower than other formats AVIF is built on advanced video codec compression designed to maximise compression ratio rather than encode speed, so it leans on more expensive prediction modes, larger transforms, and more complex filters than JPEG or WebP. The encoder here is a software build running on your device. A software AVIF encode on a typical web-sized image takes around 250 milliseconds once the encoder is warm on Chromium desktop, and around 1.1 seconds cold including the warm-up. At 8 megapixels, the encode takes around 2.8 seconds on Chrome and around 31 seconds on Firefox, which runs this kind of code more conservatively. On mid-range mobile, expect roughly 3 to 5 times the desktop figure. That is exactly why the interface never calls AVIF conversion instant. It is fine for single-photo work, but slow enough that showing a progress state is the honest thing to do. ### What happens when you pick AVIF output AVIF is the most demanding format this tool produces, so to give you the smallest, best-looking file RoundCut may run the encode on our server using a fast native build, and converting several at once always runs there. The files involved are deleted within about 2 hours, and you can clear them right away. When that path is unavailable, a compiled AVIF encoder loads once on your device, and the conversion finishes locally as a fallback. For the PNG, JPG, and WebP pairs none of this applies, since those lean on the browser's native codecs. The extra work is the price AVIF charges for its size advantage. ### How AVIF beats JPEG at the same quality AVIF uses block-based prediction much like modern video codecs. For each block of image data, the encoder tests several prediction modes, directional, smooth, and DC, against already-encoded neighbours, then keeps whichever leaves the smallest residual. Only that residual difference is transformed and quantised, not the raw image data. This captures local image structure far more efficiently than JPEG's fixed-block model, which applies the same transform geometry everywhere regardless of content. On smooth gradients, AVIF's prediction can reproduce a region with almost no residual at all. On skin tones, hair, and sky, AVIF consistently turns out smaller files than JPEG or WebP at visually equal quality. The trade-off is encode complexity, since testing many prediction modes per block is expensive, which is why AVIF takes seconds while JPEG takes milliseconds. ### Core Web Vitals: converting hero images to AVIF Largest Contentful Paint measures how fast the biggest visible element in the viewport loads. On most landing and product pages, that element is a hero photo. Google has confirmed Core Web Vitals are a ranking signal in Search. A typical hero JPEG on a 3G connection might take about 2.4 seconds to load. The same image as AVIF at the default quality could be two-thirds smaller, trimming the LCP contribution to around 0.8 seconds on the same link. That moves LCP out of the red zone above 2.5 seconds and into the green zone below it, a meaningful shift for Search performance. The saving compounds across every image on the page. CDNs with automatic format negotiation, like Cloudflare, Fastly, and Akamai, read the browser's Accept header and serve AVIF to compatible browsers on their own once AVIF originals exist in the library. ### AVIF in 2026: support, readiness, and the gaps As of mid-2026, AVIF decodes in Chrome 85+, Firefox 93+, Safari 16.4+ on iOS 16 and macOS Ventura, and Edge 121+. Combined global coverage is roughly 94 percent of browser traffic. The 6 percent gap is mostly older iOS devices still on iOS 15, which Apple has stopped updating for most modern features, plus Internet Explorer. For broad consumer audiences, AVIF is production-ready for web delivery. The remaining gaps are on the tooling side. Lightroom added AVIF export in version 13.3 in 2024, Affinity Photo has supported AVIF import and export since version 2.3, and most stock agencies still do not accept AVIF for submission. Print services universally still want JPEG or TIFF. For the web the format is ready, but for any workflow that passes files through third-party systems, keep a JPEG master in reserve. ### File sizes compared: AVIF, WebP, and JPEG Our benchmark across three real-world test fixtures gives a reliable comparison at the default quality across formats. For a typical small photo as JPEG, the WebP equivalent measured 57 percent lighter, and the AVIF measured 65 percent lighter. For a large high-resolution photo, the WebP equivalent measured 72 percent lighter, and the AVIF measured 86 percent lighter. The AVIF lead over WebP grows on large, high-resolution photos. On very small images, AVIF's advantage shrinks and the encode overhead becomes a bigger share of the total time. These are measured numbers from our own pipeline, not vendor marketing. Individual photos will vary with content complexity, noise, and colour range. --- # Convert JPG to PNG Source: https://roundcut.app/jpg-to-png/ > Re-wrap a JPG as PNG for exact-quality editing and tool compatibility. Expect a larger file. RoundCut converts JPG to PNG fast. A single image is converted on your device with no upload. Convert several at once and the files go to our server, which does the work and returns a link that is deleted within about 2 hours. Drop a JPG and the browser re-encodes it as PNG in a fraction of a second, storing every detail exactly. The PNG output preserves the current state of every part of your JPG exactly, including any compression artifacts that are already there. It does not restore detail that JPEG already discarded, because no converter can recover that. Your file size will grow, with measured results showing a small JPG becoming several times larger as PNG. This is expected behavior, not an error. The conversion is the right step when you need an exact-quality format for editing, when a tool requires PNG input, or when you want to stop further quality loss on later saves. It does not add transparency to the image. ## What it does - **One image stays local** a single file is read and encoded inside the browser tab, with no network transfer - **Several at once on our server** converting two or more sends them to our server, and the result link is deleted within about 2 hours - **Exact re-encode** the PNG adds no new quality loss, and existing JPG artifacts stay exactly as they are - **Honest size expectation** the output is larger than the source JPG (roughly 3 to 7 times), which is correct - **No transparency added** the PNG is fully opaque unless the background is removed in a separate step - **Metadata removed** EXIF, IPTC, and XMP are stripped on every browser - **Convert one or several** drop a single JPG, or drop several to convert them together - **Flexible input** both .jpg and .jpeg work, and WebP or GIF are accepted as source too ## How to use it 1. **Drop or pick your JPG** — Drag a JPG onto the upload area or click to open a file picker. Files named .jpg or .jpeg both work. Convert one and it is handled entirely on your device. Add several and they go to our server to convert together. 2. **Wait for the automatic encode** — The browser writes the PNG the instant the file loads. For most photos at screen resolution, the result appears in under a second with no button to press. 3. **Note the output size** — The output PNG will be noticeably larger than the source JPG. A small JPG typically grows to several times its original file size. This is expected behavior, not an error. 4. **Download the PNG** — Click Download to save the file. The original filename is kept with the extension changed to .png. For a batch, you download a single archive, and the link is deleted from our server within about 2 hours. To convert another file, drop the next JPG onto the page. ## How to convert JPG to PNG Drop a JPG onto the upload zone, or click to pick one from your device. The conversion starts the moment the file lands, with no Convert button to press. The result appears in under a second for typical photos. When it is ready, the Download button saves the PNG using the original filename with the extension changed to .png. Each file runs from scratch with no queue. The same code path works on desktop and mobile browsers. Files named .jpg or .jpeg behave identically. WebP and GIF files are also accepted as alternative source formats if you happen to have one of those instead. ## Why your PNG file is larger than the JPG JPG reaches its compact size by discarding detail. PNG keeps every detail with no loss. Converting your JPG to PNG therefore always produces a larger file. Measured results show a small source JPG growing to several times its original file size as PNG. This happens with every PNG converter because the PNG format is defined by its exact-storage format. A larger PNG is not a higher-quality image, since the details stored are exactly those from your JPG, artifacts included. The conversion earns its place when the exact-storage format delivers a practical benefit, such as stopping further quality loss on re-saves, meeting a tool's PNG-only requirement, or preparing your file for transparency you plan to add later. ## JPG or PNG, which format fits the workflow Switch to PNG for logos, icons, screenshots with sharp text, line art, and images that need transparency or repeated editing without piling up quality loss. Keep JPG for photographs, hero images on web pages, and uploads to social platforms that re-encode on their own anyway. Converting your JPG to PNG before editing stops each later save from adding fresh artifacts. The trade is bytes against editability, where JPG wins on size and PNG wins on preserving your image quality across many saves. That is the main reason designers and photographers move photos into PNG during a multi-step workflow. For web delivery, convert back to JPG or WebP at the final export step. ## Quality, what the conversion preserves and what it cannot The PNG output is exact, so no new quality is lost in the conversion itself. What the conversion cannot do is restore quality already lost during the original JPEG compression. Any artifacts, banding, softening, or color shifts baked into the JPG are stored faithfully in the PNG exactly as they appeared. PNG simply stops further degradation from that point on. There is no quality slider, because PNG output has no quality setting. The format stores every detail exactly by definition. If you want a sharper source image, you need a higher-quality original, since no format conversion can recover detail that JPEG already discarded. ## Does this create a transparent background No. Converting a JPG to PNG re-wraps the image data without changing what is in it. The solid-color background in your JPG stays solid in the PNG. The PNG format supports transparency as a capability, but no see-through areas are created by the conversion itself. If you need a PNG with a transparent background from a JPG, use the background-remover tool after converting. That tool isolates the subject and clears the background areas, producing a PNG with real transparency rather than a flat opaque image. ## Where your file is converted RoundCut routes your conversion based on the count of files you drop. A single image runs through the browser's built-in image engine, entirely on your device, finishing in well under a second with nothing sent to any server. Two or more images are packaged and sent to our server so they can be handled in one batch, letting you download a single archive instead of saving files one by one. The server result is deleted within about 2 hours. The split exists because large batches would tax your browser's resources if processed locally, while a single file is fast and straightforward on-device. Either way, the download appears the same moment the conversion finishes. ## Questions **Does converting JPG to PNG improve quality?** No. The conversion does not restore quality that JPEG already discarded. The PNG output preserves the current state of your image exactly, including any existing compression artifacts. What converting does stop is further quality loss, since once your file is in PNG you can edit and re-save without adding new compression damage on each pass. The PNG output is exact, but the JPG source was not. For a sharper starting point, you need a higher-quality original file. **Why is my PNG file larger than the original JPG?** PNG stores every detail without any compression loss. JPG reaches its compact size by permanently discarding detail it considers imperceptible. PNG keeps all of it. A small JPG typically becomes several times larger as PNG, with a large photo growing even more. This is expected and happens with every PNG converter. If smaller files matter more than an exact-quality format, stay on JPG or use a compress tool after the conversion. **Does converting JPG to PNG create a transparent background?** No. Converting changes the image format but not your image content. The solid background in your JPG remains solid in the PNG. PNG supports transparency as a feature, but no see-through areas are added by the conversion itself. To produce a PNG with a transparent background, you need to remove the background separately using a dedicated background-remover tool. **Is it safe to convert JPG to PNG here?** It depends on how many files you convert. A single image is converted on your device with nothing uploaded, which you can confirm by watching your network traffic during the conversion and seeing no outbound image request. Convert several at once and they are sent to our server to be processed, then the download link and the files behind it are deleted within about 2 hours. If your images are sensitive, convert them one at a time to keep everything on your device. **Can I convert multiple JPG files to PNG at once?** Yes. Drop several JPG files and RoundCut converts them together and hands back a single archive to download. A batch is processed on our server rather than on your device, and the download link is deleted within about 2 hours. A single file, by contrast, is converted on your device with nothing uploaded. Each conversion is fast either way. **Why would I convert JPG to PNG?** Three main cases. You need an exact-copy format to edit your image several times without adding fresh artifacts on each save, a tool or workflow requires PNG input and will not accept JPG, or you want to add transparency later and need PNG's transparency support. None of these cases restore JPG quality, they just stop the quality from getting worse. If you only need a smaller file, a compress tool is the better choice. ## The details ### The exact-storage format, what PNG stores PNG uses a no-loss compression algorithm. It stores every color value exactly as provided, applies a reversible filter to each scanline, and compresses the result. The key property is that the decompressed values are byte-identical to the originals. For the JPG-to-PNG case, the originals are the pixels produced when the browser decodes the JPEG. Those decoded pixels already reflect all the approximations JPEG made during its original encode, so PNG faithfully stores the approximated pixels, not the real-world scene. Measured, a small JPEG at medium resolution grows to several times its original file size as PNG, and a larger JPEG at high resolution grows even more. These multiples hold roughly across content types because the ratio tracks resolution, not JPEG quality level. The exact-storage format is valuable for what it prevents, further degradation, not for what it restores. ### Why JPEG quality cannot be recovered JPEG compression discards information permanently. The encoder breaks the image into small blocks and stores a simplified version of each, rounding fine detail down to a smaller set of values. That rounding is one-way: detail dropped during encoding cannot be recovered later, and no record of the original survives in the file. When the browser opens the JPEG, it rebuilds the picture from those rounded values, which are approximations of the original. Saving those approximated values as PNG produces an exact record of the approximation, so the PNG is a perfect copy of the already-degraded image. This is not a limit of PNG or of this tool. It is a basic property of compression that throws data away: what is dropped at save time is gone. To improve quality you have to start from an uncompressed or RAW original. ### Measured file size growth The size ratio from JPG to PNG varies by image content but follows a predictable pattern. Photographic images with complex tonal variation grow the most, because JPEG's transform is tuned for exactly that content and achieves high compression ratios, while PNG's exact-storage compression cannot match those ratios on noisy image data. Test measurements from this tool show a small JPEG photo at medium resolution growing roughly 6 times as PNG, and a larger JPEG at high resolution growing roughly 3.3 times. For flat-color images like screenshots and icons, JPEG is already poorly suited to the content and its files tend to be larger for equivalent quality, so the PNG of the same content grows less dramatically. The practical implication is direct, if output file size matters for your use case, converting a JPG to PNG makes the situation worse, not better. ### Transparency, the capability versus the content PNG supports full transparency as a built-in feature, where a file can include per-area opacity values ranging from fully transparent to fully opaque. When a JPG is converted to PNG, the resulting PNG is set to fully opaque, because the source JPG had no transparency information to begin with. The PNG format is ready to hold transparency data. The file just does not contain any, because none was present in your source. Adding transparency to the image requires separate processing, either masking the background in an editor or using an automated background-removal step. A background-remover tool trained to identify the subject can produce a PNG with real transparency by clearing the background areas after the conversion. ### EXIF metadata handling The re-encode pipeline strips EXIF, IPTC, and XMP metadata from the PNG output on every browser. This means GPS coordinates, camera model, capture date, copyright strings, and any custom XMP fields present in the source JPG are removed. ICC color profiles follow a slightly different path, where Chrome and Safari preserve the sRGB ICC profile tag in the output and Firefox strips it along with all other metadata. The practical result is sRGB-safe output across browsers, but wide-gamut profiles such as Display-P3 or Adobe RGB are lost in Firefox. For most web and sharing uses, metadata removal is helpful, since it reduces output size slightly and removes location data from photos. For professional photography or archival workflows where embedded metadata must be preserved, handle the metadata chain with a dedicated tool before or after the format conversion. ### Privacy and where conversion happens Where the work happens depends on the number of files. For a single image, no data leaves your device, and that is verifiable in real time. You can verify this by watching your browser network panel during a single conversion and confirming zero outbound requests carry image data. The JPEG decode and PNG encode both happen on your device, off the network. For two or more files, RoundCut sends them to our server, which converts them, zips the result, and returns a download link. That link and the converted files are deleted within about 2 hours. The single-image path keeps everything on your device, while the batch path trades that for converting many files in one step. When your image content is sensitive, converting one at a time leaves no upload at all. --- # Convert JPG to WebP, Smaller Files in Seconds Source: https://roundcut.app/jpg-to-webp/ > Turn any JPEG into a lighter WebP and cut file size by 25 to 34 percent. Drop one file or several at once. Convert JPG to WebP right from the browser tab. Drop a JPEG and your browser re-encodes it as WebP, then hands you the download in about a second, with a single image converting on your own machine and nothing uploaded. WebP compresses photos more cleverly than JPEG, so files usually come out 25 to 34 percent lighter than the source, and our own tests confirm that figure on a range of real photos. At the near-exact quality used here the visual difference is invisible on photographs at any normal viewing distance. Lighter images load faster, which helps your Largest Contentful Paint score. Want to convert several at once? Drop them together and they go to our server to process as a batch, where the download link is removed within about 2 hours. ## What it does - Native WebP output. Your browser already speaks WebP, so no extra module downloads for a JPG to WebP job. - 25 to 34 percent lighter. Google publishes that range for WebP versus JPEG at equal quality, and our own tests confirm it. - One image converts on your machine. Convert several at once and they go to our server, where the download link is removed within about 2 hours. - Tuned for near-exact quality. The quality target sits where JPEG and WebP look identical on photographs. - Zero added page weight. WebP output rides the browser codec, so the page ships no extra bytes for this conversion. - Metadata dropped. The output WebP carries no EXIF, no GPS, and no camera fingerprint, only the visible detail. - Auto-renamed download. The file keeps its original name and simply gains the .webp extension. - Runs after first load. Once the page is cached the conversion works without a connection. ## How to use it 1. **Drop or pick your JPG** — Click the upload area or drag a JPEG onto it from your desktop. The conversion starts on its own the moment the file lands, with no button to press. 2. **Let the encode run** — Your browser re-encodes a single photo into WebP in its own memory, with nothing to upload. A typical phone shot wraps up in well under a second. 3. **Read the size comparison** — When the WebP is ready, a stats line shows the old JPEG size and the new WebP size side by side so you can confirm exactly how much you saved. 4. **Download the WebP** — Click Download to save the result. The file keeps its original name and switches to the .webp extension for you, so nothing needs renaming. ## How to convert JPG to WebP Drag a JPEG onto the upload area, or click it to pick a file from your device. The conversion fires the instant the file lands. There is no Convert button to hunt for and no queue to wait through. Your browser re-encodes the photo as WebP entirely on your own hardware. When the result is ready, the stats line puts the original size and the new size next to each other so the saving is plain to see. Click Download to keep the WebP, which carries the same base name with a fresh extension. To do another photo, just drop it on top of the last one. The page never reloads, and a typical phone shot finishes in well under a second. ## Why WebP comes out smaller than JPG Both formats throw away data to shrink a photo, but WebP does it more cleverly at any given quality. Google puts the typical gain at 25 to 34 percent over JPEG, and busy real-world photos often stretch past that. Our measured runs back it up. A compact JPEG phone photo dropped to under half its original size when converted to WebP. Fewer image bytes mean a quicker Largest Contentful Paint, the loading metric Google folds into its ranking systems. WebP is also open to use, which is exactly why every big CDN serves it by default and every current browser reads it without a plugin or a download. ## When WebP beats JPG and when it does not Reach for WebP on anything bound for a web page you run: hero shots, product photos, blog art, gallery thumbnails, and the preview cards your own domain serves. Every modern browser reads it, which covers more than 97 percent of traffic in 2026, Safari included since version 14. Stay on JPG when the file has to land somewhere that still expects JPEG, such as an email attachment, a print job, an older photo editor, a marketplace upload form, or a platform that quietly re-compresses WebP back to JPEG anyway. The destination decides, not the starting point. If you control where the image gets served and your audience is on recent browsers, WebP is the stronger pick in nearly every case. ## Quality and the second lossy pass Your JPEG was already compressed, so turning it into WebP stacks a second round of compression on top of the first. On paper, extra artifacts could pile up. In practice, at the near-exact quality setting used here, photos look the same as the source at any normal viewing distance. For the cleanest output, always start from the best JPEG you own, an original off the camera rather than a copy that already passed through a chat app or a CDN squeeze. Detail loss adds up with every hop. This version ships one calibrated quality level with no slider, which keeps the output predictable and removes the temptation to over-compress a photo that did not need it. ## Where your file is handled Converting one image happens right on your own machine, with nothing uploaded. If you want to check, open the Network panel in the browser developer tools and convert a single photo. You will see no outbound image request. When you convert several at once, the files go to our server to do the work together, and the download link is removed within about 2 hours. Either way the result is yours to keep, and nothing is logged or studied. ## WebP browser support in 2026 WebP reads in more than 97 percent of browsers worldwide today. Chrome has decoded it since version 17 back in 2012, Firefox since version 65, and Safari since version 14 on iOS and version 16 on macOS Ventura. Edge and Opera both handle it natively. The only real holdouts are Internet Explorer 11 and Safari 13, and their share is now tiny. For general web delivery you can serve WebP with confidence. If you genuinely must reach ancient browsers, pair the WebP with a JPG fallback inside an HTML picture element and let each browser pick what it can read. For email, shared drives, and print, JPG stays the safer pick. ## Questions **Why convert JPG to WebP?** WebP makes smaller files than JPEG at the same visual quality, which speeds up page loads and lifts your Core Web Vitals. Google's own figure for WebP is 25 to 34 percent lighter than JPEG, and real photos often do better. If you manage images for a site, moving from JPEG to WebP trims bandwidth bills and can directly improve your Largest Contentful Paint, which feeds into Search ranking. **Is this conversion lossless?** No. WebP and JPEG are both compressed formats, so going from one to the other adds a second round of compression. At the near-exact quality used here the difference is invisible on photos at normal viewing distances. If you truly need a result that captures every detail without any loss, convert to PNG instead. Any tool that promises an exact JPEG to WebP conversion without any loss is wrong, since that mode is a separate WebP type meant for simple graphics, not photographs. **How much smaller are WebP files than JPEG?** In our tests a compact JPEG phone photo dropped to under half its original size when converted to WebP. A larger photo at high resolution fell to roughly a quarter. Google's published figure is 25 to 34 percent lighter than JPEG at the same quality. Very small images that are already heavily squeezed, like tiny thumbnails, tend to gain less because they sit close to the encoder's floor already. **Does WebP support transparency?** Yes. WebP carries full transparency while JPEG does not, which is one reason it suits web graphics. For a JPG to WebP job, though, there is no transparency to keep, since the source JPEG is fully opaque to begin with. If your image started life as a PNG with transparency and you want that preserved in WebP, convert straight from the PNG rather than from a flattened JPEG in the middle. **Is my image uploaded to a server?** It depends on how many. Converting one image runs entirely on your machine, with no outbound image request. You can confirm this by watching the Network panel in the browser developer tools. Converting several at once sends them to our server so they process together, and the download link is removed within about 2 hours. There is no long-term storage and no data collection either way. **Can I convert several JPGs to WebP at once?** Yes. Drop two or more JPEGs and they convert together, then come back as one download. To do this the files go to our server, which packs the WebP results into one archive and gives you a single link that is removed within about 2 hours. A single JPEG still converts right on your machine with nothing uploaded. ## The details ### How a single image converts on your machine, and what changes for several When you drop one JPEG on the page, the browser reads the file into memory, decodes the compressed data back into raw detail with its built-in JPEG reader, draws those results into an off-screen surface, and asks the platform to write them back out as WebP. Every step happens right on your machine, so no bytes cross the network. In our runs a compact phone JPEG finishes in roughly 54 milliseconds, and a large high-resolution photo takes about 550 milliseconds. When you convert several images at once, the work moves to our server instead, which encodes them together and returns one download whose link is removed within about 2 hours. ### What happens to EXIF and metadata JPEG files can carry real metadata payloads. EXIF records the camera model, GPS coordinates, shutter speed, and orientation. IPTC fields hold copyright and caption. XMP packets track edit history, and ICC profiles cover colour-managed work. The conversion here keeps none of it. The output WebP is a clean file holding only the visible detail. That is the normal behaviour for image tools that draw through the browser, across Chromium, Firefox, and WebKit alike. For most web publishing, dropping the metadata is a plus: it trims a few bytes and stops GPS or personal data leaking into a public image. If you need one specific field kept, orientation being the usual one, rotate the photo first or use a metadata-aware editor before converting. Do not lean on this tool for archival work where the original metadata has to survive intact. ### WebP versus JPEG: how the compression differs JPEG slices an image into small square blocks and runs a Discrete Cosine Transform to turn spatial detail into frequency data, then compresses the higher frequencies harder. WebP borrows a prediction scheme from VP8 video: each block is guessed from its already-decoded neighbours, and only the leftover difference gets stored. That tends to leave fewer blocky artifacts at the same file size, especially across smooth gradients and skin tones where JPEG's grid can surface as a visible mosaic. On photos at high quality, the eye rarely catches the difference, but the size advantage holds steady. At lower quality, WebP's lead widens and the artifact character shifts: WebP smears into a softer blur while JPEG shows its trademark blocks. Neither wins on every image. Heavily textured subjects like foliage sometimes compress about the same in both formats. ### Core Web Vitals and the case for WebP Largest Contentful Paint, or LCP, measures how fast the biggest visible element finishes loading in the viewport. On most marketing pages that element is a hero photo. Google has confirmed Core Web Vitals act as a ranking factor in Search. When the hero is a large JPEG, it can shove LCP past the 2.5 second threshold on a mid-range mobile connection. Convert the same photo to WebP and LCP can slide back inside the green zone with no other change to the page. The 25 to 34 percent saving compounds across every image on the page. A page carrying several JPEGs can shed hundreds of kilobytes by switching to WebP, which directly speeds up the first meaningful paint on slow links. CDNs that support content negotiation hand WebP to compatible browsers on their own, so there is no per-image busywork once the originals exist. ### Browser support for WebP in 2026 and the gaps As of 2026, WebP reads in Chrome since version 17, Firefox since 65, Edge since 18, Opera since 11.10, and Safari since version 14 on iOS 14 and macOS Big Sur. According to caniuse.com, combined global coverage for WebP decode tops 97 percent of browser traffic. The remaining sliver is mostly Internet Explorer 11, Safari 13 on macOS Catalina, and a long tail of very old Android browsers. For most public projects, serving WebP to everyone is safe. When you need total coverage, the HTML picture element lets you list a WebP source and a JPEG fallback in one tag, and the browser takes the first format it can read. CDNs with image optimisation negotiate the format automatically from the Accept header the browser sends, so no manual per-browser testing is needed when you serve through them. ### When not to use WebP WebP is the right call for web delivery to modern browsers, but it is the wrong call in a few common spots. Print work needs CMYK colour, which WebP does not carry, so JPEG and TIFF stay standard for the press. Email clients are uneven: Gmail and Apple Mail render WebP, but Outlook on Windows does not. File sharing and marketplaces vary too. Google Drive, Dropbox, and GitHub show WebP fine, while many social platforms, stock agencies, and e-commerce systems re-compress incoming images to JPEG internally, which makes the WebP step pointless. Editor support is still patchy: Lightroom, Capture One, and Affinity Photo can open WebP, yet plenty of plugins and export presets still default to JPEG. For any workflow that passes images through third-party systems you do not control, keep a JPEG master and convert to WebP only at the final delivery layer. --- # Convert PNG to AVIF, Smallest Files With Alpha Kept Source: https://roundcut.app/png-to-avif/ > Drop a PNG and get an AVIF that keeps your alpha channel while cutting file weight on photographs. Convert a PNG to AVIF with full transparency carried through. Drop the file onto the page and your PNG is decoded, then re-encoded as AVIF, a modern image format built for perceptual efficiency. Because AVIF encoding is heavy, RoundCut runs it on our server for the best result, and if that path is unavailable the conversion falls back to an in-browser encoder. The output is encoded at quality 85, visually near-indistinguishable from the source and typically 30 to 50 percent smaller than the same image as WebP. Converting to JPG would flatten transparency, but AVIF carries the whole transparent layer intact. When a file is sent to our server it is processed and the result is removed within about 2 hours. This pair is built for developers and designers who want the smallest modern format for transparent assets. ## What it does - Transparency preserved, every transparent area in your PNG stays transparent in the AVIF, with no flattening and no white fill. - AVIF is encoded on our server for the best result, with an in-browser encoder as an automatic fallback if the server is unavailable. - Typically 30 to 50 percent smaller than WebP on photographs at comparable visual quality, and far smaller than the PNG itself. - Encoded at quality 85, around 42.6 dB PSNR on natural images, so fine photographic detail holds up well. - Honest encode time, a 4K photo takes about 2.8 seconds on Chrome and warm encodes after the first are quicker. - EXIF, IPTC and XMP metadata are stripped, which helps privacy but also drops orientation data. - One PNG in, one AVIF out, single file per conversion in this version. - Files sent to our server are processed and the result is deleted within about 2 hours, with nothing kept beyond that. ## How to use it 1. **Drop your PNG** — Drag the PNG onto the upload area, or click the area to open a file picker and choose one from your device. 2. **Wait for the encode** — The first conversion loads the AVIF encoder in about a second. Later conversions in the same tab run faster without that warm-up. 3. **Check the sizes** — When the AVIF is ready, the stats line shows the original PNG size next to the new AVIF size so you can confirm the reduction. 4. **Download the AVIF** — Click Download to save the file to your device with the same base name and a new .avif extension. ## How to convert PNG to AVIF Drop a PNG onto the upload area or click it to pick a file. The conversion starts the moment the file lands, with no separate convert button. The first encode of a session loads the AVIF encoder in about a second. After that, a small image converts in roughly 40 milliseconds, a 1-megapixel photo in about 250 milliseconds, and a 4K shot in around 2.8 seconds. Images above 8 megapixels can take 10 to 30 seconds on a phone. When the AVIF is ready, click Download to save it with the same base name and a new .avif extension. The stats line reports both file sizes so the reduction is visible at a glance. ## Does AVIF preserve transparency? Yes, and this is the main reason to choose PNG to AVIF over PNG to JPG. AVIF supports transparency natively, so every transparent area in your source survives the conversion unchanged. A logo with a soft drop shadow, a product cut-out with feathered edges, an icon sprite with rounded corners, all arrive in the AVIF with the same masking, no flattening, no painted-in background. JPG has no transparency at all and replaces it with a solid color. WebP also keeps transparency, but AVIF reaches a smaller file at comparable quality. For transparent assets headed to a modern browser, AVIF is the most efficient option available. ## How much smaller is AVIF than PNG or WebP? At quality 85, AVIF is typically 30 to 50 percent smaller than PNG on photographs and 20 to 30 percent smaller than WebP. In a development test, a 4K photograph that weighed 116 kilobytes as JPG dropped to 16 kilobytes as AVIF. Smaller images follow a similar ratio, a 17-kilobyte PNG converting to roughly 6 kilobytes. Heavily compressed sources narrow the gap, and graphics with sharp edges or large flat color areas may shrink less. The biggest savings land on natural photographs, which is the category that dominates page weight on most product and landing pages today. ## Quality and encode speed AVIF compresses with some quality loss, and this pair encodes at quality 85. At that setting, photographs measure around 42.6 dB PSNR, visually indistinguishable from the PNG for natural images. Graphics with sharp text or fine line art can show subtle artifacts at any compression setting, so keep the PNG for those. Speed depends on size and browser. On Chrome desktop, 1 megapixel takes about 250 milliseconds and 8 megapixels roughly 2.8 seconds. Firefox runs the encoder about four times slower, and a phone adds another 3 to 5 times on top. The first conversion of a session pays a one-time warm-up while the encoder initializes. ## Where your file is processed AVIF encoding is demanding, so RoundCut runs this conversion on our server to get the best quality and speed, and falls back to an in-browser encoder if the server cannot be reached. That means your PNG may be sent to us when you convert it. The file is encoded to AVIF, the result is returned to you, and both are removed from our server within about 2 hours. We do not keep your image, we do not ask for an account, and we do not use it for anything other than the conversion you requested. If the server cannot be reached, the conversion falls back to an encoder that runs locally instead. Either way you get the same AVIF, with the same transparency and the same quality setting. ## AVIF browser support in 2026 AVIF reaches about 94.3 percent of global browsers in 2026: Chrome 85 and later, Firefox 93 and later, Safari 16.4 and later (iOS 16 and later), and Edge 121 and later. Internet Explorer, Opera Mini, and iOS 15 or older do not display it. If your audience includes those, serve AVIF first and let the HTML picture element fall back to WebP so unsupported browsers get the next-best option automatically. For most modern audiences the fallback case is small, since Chrome and Safari cover the bulk of mobile traffic and both shipped native AVIF support back in 2022. ## Questions **Why convert PNG to AVIF?** AVIF gives the smallest file of any mainstream browser-supported format while keeping transparency intact. For developers and designers, PNG to AVIF is the right call for transparent UI assets, product cut-outs, and graphic icons that need to stay light on bandwidth. The saving is real and measurable, typically 30 to 50 percent under the PNG at visually similar quality, with no flattening of transparency that JPG cannot keep. **Does AVIF support transparency?** Yes. AVIF supports transparency natively, and this pair carries it through. Every transparent area in your PNG stays transparent in the AVIF. Unlike JPG, which has no transparency and replaces it with a solid color, AVIF handles cut-outs, logos, and layered assets exactly as PNG does, just at a much smaller file size. No manual masking is needed at any point. **Which browsers display AVIF?** Chrome 85 and later, Firefox 93 and later, Safari 16.4 and later (including iOS 16 and later), and Edge 121 and later all display AVIF natively. Internet Explorer, Opera Mini, and devices still on iOS 15 or earlier do not. If your deployment needs to cover those, use the HTML picture element to serve AVIF to supporting browsers and WebP or PNG as the fallback. **Is AVIF quality exact or compressed?** AVIF can be encoded either way, but this pair encodes with compression at quality 85. Photographs at that setting measure around 42.6 dB PSNR, which is visually near-exact to most viewers. Graphics with hard edges, fine text, or flat color may show subtle compression artifacts at any quality. If you need an output that preserves every detail exactly, keep the original PNG instead of converting. **How does AVIF compare to WebP?** At comparable visual quality, AVIF is roughly 20 to 30 percent smaller than WebP on photographs. WebP has broader support, including older Safari versions, while AVIF reaches about 94.3 percent of global browsers in 2026. For modern audiences where every kilobyte counts, AVIF is the better default. For maximum compatibility, keep WebP as a fallback. The 2026 pattern is AVIF first with a WebP fallback through the picture element. **How long does PNG to AVIF conversion take?** It depends on size and browser. The first conversion of a session loads the encoder in about a second. After that, a small image takes around 40 milliseconds, a 1-megapixel photo about 250 milliseconds, and a 4K photograph roughly 2.8 seconds on Chrome desktop. Images above 8 megapixels on a phone can take 10 to 30 seconds. Firefox is about four times slower for AVIF, so Chrome gives noticeably faster results on large files. ## The details ### Why AVIF beats PNG so decisively on file size AVIF is built on a modern codec designed for perceptual efficiency rather than exact reproduction. PNG uses DEFLATE on raw color data, which is exact but leaves enormous redundancy intact in photographic content. AVIF replaces that redundancy with block prediction borrowed from video, discarding information the human visual system does not notice. The result is dramatic. A 116-kilobyte photo converts to roughly 16 kilobytes as AVIF at quality 85, and a transparent PNG usually shrinks 30 to 50 percent against an equivalent WebP. The gap widens with high-resolution photographs and narrows on simple graphics. For any site serving transparent images to a modern audience, the bandwidth saving translates straight into faster loads and lower egress cost. ### The alpha channel story in detail Both PNG and AVIF store transparency as a separate layer alongside the color data. When this pair processes your PNG, the decoder extracts the color data and the transparency mask independently. The AVIF encoder then writes a new file with its own transparency track, compressing both layers at quality 85. The transparency is not baked into the color and not replaced by a background fill. Soft shadows, feathered edges, and partially transparent gradients are all encoded faithfully. The one change AVIF introduces is compression on the transparency layer itself, which can create barely perceptible fringing around very hard edges at extreme zoom. At normal display sizes and quality 85, the difference is not visible. For precise icon work at tiny sizes, keep the PNG. ### Encode speed in practice, browser by browser The AVIF encoder loads once per browser session, about 800 milliseconds to fetch plus 300 to initialize, so the first conversion carries roughly a second of warm-up. Warm encodes skip that entirely. On Chrome desktop, 0.12 megapixels takes about 40 milliseconds, 1 megapixel around 250 milliseconds, and 8 megapixels approximately 2.8 seconds. Firefox is the outlier, running the same encoder about four times slower, so a 4K photo can take 30 seconds or more there. WebKit on Safari sits between the two, closer to Chrome. On mid-range mobile hardware the numbers run three to five times worse than desktop Chrome. If you convert large files often, Chrome on a desktop or laptop gives the best throughput. ### When to keep PNG instead of converting PNG stays the right choice in several cases even when file size matters. First, graphics with sharp text at small sizes, like labels, badges, or favicon-scale icons, can show artifacts in AVIF that are unacceptable at typical viewing distances. Second, source-of-truth files used for further editing should stay PNG, since every pass through a compressed format compounds quality loss. Third, environments that reject AVIF, including some document editors, older design tools, and email clients, need PNG for interoperability. Finally, animation beyond the first frame is not preserved, since this pair handles a single frame. For everything else, transparent photographs and graphic assets headed to modern web delivery, PNG to AVIF is the right trade. ### Core Web Vitals and the bandwidth argument Largest Contentful Paint, the primary loading metric, is directly affected by the size of the largest image on a page. Reducing that image from 116 kilobytes to 16 kilobytes, as measured on a 4K photo in development, cuts transfer time on a 10 Mbps connection from roughly 93 milliseconds to roughly 13 for that element alone. Multiply that across a product grid, a hero carousel, or a UI packed with transparent icons and the cumulative saving pushes LCP well under the 2.5-second threshold Google treats as good. AVIF adoption accelerated precisely because the argument can be made in hard numbers. At about 94.3 percent browser coverage, the fallback case is small enough that the bandwidth win outweighs the extra picture element markup. ### How this differs from a typical online converter Most online AVIF converters upload your PNG to remote machines, process it there, and hold it under whatever retention policy the provider sets, which is often vague. RoundCut is direct about its flow. Because AVIF encoding is heavy, the conversion runs on our server for the best result, with an in-browser encoder as an automatic fallback when the server is unavailable. When your file is sent to us, it is encoded and then removed within about 2 hours, without asking for an account and with no use beyond the conversion you asked for. There is no permanent storage and no sharing of your image. For developers working with client documents, proprietary product photography, or user-generated content, the honest version of the trade is this: the file may pass through our server, it is handled only to produce your AVIF, and it does not stick around. --- # Convert PNG to JPG Source: https://roundcut.app/png-to-jpg/ > Drop PNG to JPG for photos, email attachments, and uploads where smaller files matter. RoundCut converts PNG to JPG in under a second for typical photos. A single image is converted on your device with no upload. Convert several at once and they are processed on our server, then the download link is removed after about 2 hours. Drop a PNG and the re-encode finishes quickly. Transparent areas of your PNG are replaced by a solid white background, because JPG cannot carry transparency. For photographic content, the result is usually 30 to 50 percent smaller than the source PNG at a quality the eye cannot tell apart. On a measured test, a photo PNG produced a matching JPG in 10 ms at small sizes, and a larger photo PNG converted in about 1.5 seconds. EXIF, IPTC, and XMP metadata are stripped in every case. This conversion is the right choice when file size matters and your image is a photograph. For logos, icons, or anything with transparency you want to keep, choose PNG to WebP instead. ## What it does - **One image stays local** a single file is read and encoded inside the browser tab, nothing to upload - **Several at once on our server** converting two or more sends them to our server, and the result link is deleted within about 2 hours - **Fast encode** a photo PNG converts in about 10 ms on desktop (measured) - **Transparency flattens** transparent pixels are filled with white, the expected JPG behavior - **Metadata removed** EXIF, IPTC, and XMP are stripped on every browser - **Works on photos** photographic PNGs usually shrink 30 to 50 percent at near-identical quality - **Honest about loss** no false quality claim, just an accurate JPG re-encode - **Simple flow** drop one PNG and download the JPG, or drop several to convert them together ## How to use it 1. **Drop or select your PNG** — Drag a PNG file onto the upload area, or click the area to open a file picker. Convert one image and it is handled entirely on your device with no upload. Add several and they are sent to our server to convert together. 2. **Wait for the automatic conversion** — The browser encodes the JPG the moment the file lands. There is no button to click and no progress bar for standard photos, the result appears in under a second. 3. **Check the output size** — The Download button shows the output file size. Any transparent areas now appear with a white background in the JPG, which is expected since JPG has no alpha. 4. **Download the JPG** — Click Download to save the file. The original filename is kept with the extension changed to .jpg. For a batch, you download a single archive, and the link is deleted from our server within about 2 hours. To convert another PNG, drop the next file onto the page. ## How to convert PNG to JPG Drop a PNG onto the upload zone, or click to pick a file from your device. The conversion starts the moment the file lands and usually finishes in well under a second for photos at normal screen resolution. When the result is ready, the Download button saves the JPG using the original filename with the extension swapped. To convert another image, drop the next file. Each conversion runs from scratch with no queue and no warm-up cost. The same code path works on desktop and mobile browsers, and an animated PNG is processed as a single still frame from the first image in the sequence. ## What happens to transparent areas JPG cannot carry transparency. Every transparent area of your PNG has to be filled with a solid color before the file can be encoded as JPG. RoundCut fills those areas with white, the standard behavior and the reason logos and product cut-outs end up with a white box after conversion. It is not a bug, it is the format. If you need a different background color, paint it onto the PNG in an image editor before converting. If you want to keep the transparent cut-out for web use, convert to WebP instead. WebP carries full transparency support and usually produces a file smaller than the PNG without flattening anything. ## PNG or JPG, which format fits your image Keep PNG for logos, icons, screenshots with sharp text or line art, and anything with transparency you plan to reuse. PNG stores every detail exactly, so edges stay crisp and you can edit without adding new compression noise on each save. Switch to JPG for photographs, banner images, and anything headed to a social platform that re-encodes on upload anyway. A photographic PNG is far larger than it needs to be for sharing, and JPEG was designed for that content. If you want modern-format benefits like smaller files with alpha intact, WebP covers both goals and is supported in every current browser. ## Quality settings and file size trade-offs JPG uses compression that discards some detail to reach a smaller file. For photographs, the visible difference at typical web quality is negligible. The encode target sits around 43 dB PSNR, the level where the eye cannot distinguish a JPG from the source on photographic content. Graphics with hard edges, fine text, solid color blocks, or gradients that snap to a single tone behave differently, since JPG introduces ringing around those transitions at any quality level. For those images, PNG is the correct format and converting to JPG is a downgrade. There is no quality slider here, the setting is fixed at the near-identical level calibrated for photographs. ## Where your file is converted Two paths exist, depending on how many files you drop. One PNG is handled entirely on your device: the encode runs on your device, nothing is uploaded, and the result downloads directly from the page. That path is instant and leaves no trace on any server. Drop two or more PNGs at the same time and the files travel to our server instead, which converts them together, packages the result, and returns a download link. That link and the converted files are removed within about 2 hours, and you can delete the result yourself the moment you have saved it. Choose whichever path matches how sensitive your images are. ## When to keep the PNG instead Converting a PNG to JPG is the wrong move in a few clear cases. If the PNG has a transparent background you need to preserve, JPG will destroy it, so use WebP or keep the PNG. If the image contains fine text, sharp lines, a logo, or a UI screenshot, JPG adds visible ringing at any quality level because its block-based algorithm struggles with hard edges. If you intend to edit the file further and re-save it several times, each JPG re-encode adds fresh loss, so keep the PNG as the working master and export to JPG only at the final delivery step. ## Questions **Does converting PNG to JPG reduce quality?** Yes. JPG discards some detail to reach a smaller file. RoundCut uses a fixed quality setting calibrated for photographs, where the output looks identical to the source PNG at normal viewing sizes. The encode targets around 43 dB PSNR, a level the eye cannot distinguish from the source on photographic content. Graphics with sharp edges, thin text, or flat color blocks may show subtle artifacts at any JPG quality. For those images, keeping your PNG is the correct choice. **What happens to transparent areas when I convert PNG to JPG?** JPG cannot carry transparency. The conversion fills every transparent area in your image with white, which is why logos and product images on transparent backgrounds get a white box afterward. This is expected format behavior, not a tool defect. To avoid it, either paint the desired background color onto your PNG before converting, or switch to WebP, which preserves transparency and usually produces a smaller file than PNG. **How much smaller will the JPG be?** For photographic PNGs, the JPG is usually 30 to 50 percent smaller at the near-identical quality setting used here. The exact ratio depends on the image, with high-detail photos gaining the most. Flat-color graphics, icons, and line art shrink less and may look worse because of JPG artifacts. If the PNG started as a screen capture or a logo, JPG may offer no meaningful size advantage and could reduce quality visibly. **Is it safe to convert PNG to JPG here?** It depends on how many files you convert. A single image is converted locally on your device with nothing uploaded. Convert several at once and they are sent to our server to be processed, then the download link and the files behind it are deleted within about 2 hours. If your images are sensitive, convert them one at a time to keep everything on your device. **Can I convert multiple PNG files to JPG at once?** Yes. Drop several PNG files and RoundCut converts them together and hands back a single archive to download. A batch is processed on our server rather than on your device, and the download link is deleted within about 2 hours. A single file, by contrast, is converted on your device with nothing uploaded. Each conversion is fast either way. **When should I keep the PNG instead of converting to JPG?** Keep the PNG when the image has transparency you need to preserve, when it contains sharp text or line art where JPG artifacts would show, or when you plan to keep editing and re-saving it. Each time a JPG is opened and re-saved, new compression loss accumulates. For working files, PNG is the right master format, and you export to JPG only at final delivery. For web assets that need transparency alongside a smaller file, WebP is the better export. ## The details ### Why photographic PNGs are so large PNG compression stores every dot of color exactly as it appears. For a photographic image with millions of subtly varying tones, that produces very large files because there is almost nothing to deduplicate or predict between neighboring areas. A typical 1600x1200 photo saves as a large PNG. The same image as a JPEG at quality 85 is much smaller in file size. That gap comes from the transform at the heart of JPEG, which divides the image into small blocks, discards detail the human visual system is least sensitive to, and stores the remaining data compactly. The result is a much smaller file for any image with continuous-tone variation, which is essentially every photo. PNG's exact-storage approach is valuable for graphics and working files, but it is the wrong format for photographs headed out for sharing. ### Transparency and the white background PNG lets each area of an image be fully see-through, partially see-through, or fully solid. JPG has no place to store that see-through quality at all. When the browser re-encodes a PNG to JPEG, it has to fill every transparent area with a solid color before writing the file. The default fill is white, which is why transparent areas in your PNG appear white in the resulting JPG. The key point is that no tool can produce a transparent JPG, because the format was never designed for it. If preserving the cut-out matters, the answer is WebP or keeping the PNG. RoundCut chooses white as the fill color since it matches the background of most documents and product listings. ### Measured encode performance Measured on Chrome 148, Linux desktop, using the platform JPEG encode path. A small photo (around 0.12 megapixels) encodes in roughly 10 to 15 ms. A medium photo (around 0.78 megapixels) encodes in about 13 to 20 ms. A large photo (around 8 megapixels) encodes in roughly 1.4 seconds. A very large photo (around 48 megapixels) encodes in about 1.5 seconds. JPEG encoding in the browser is substantially faster than PNG encoding for the same image, and far faster than AVIF, which needs a separate module and can take 25 seconds at maximum resolution even on desktop Chrome. This speed advantage makes PNG to JPG one of the fastest conversion paths in this family, with no extra loading cost because JPEG encoding is built into every browser. ### EXIF and metadata handling The re-encode pipeline strips EXIF, IPTC, and XMP metadata from the output JPG on every browser. That means location coordinates, camera model, capture date, copyright notice, and any custom fields present in your source PNG are removed. ICC color profiles follow a different path, where Chrome and Safari preserve the sRGB ICC profile in the output while Firefox strips it entirely. The practical result is that the converted JPG is sRGB-safe across all browsers, but if your PNG was tagged with a wide-gamut profile such as Display-P3 or Adobe RGB, that tag does not survive in Firefox. If preserving complete metadata is a workflow requirement, edit it with a dedicated tool after conversion. For typical web and sharing use, metadata removal is helpful because it trims the file slightly and drops GPS data from photos. ### When JPG artifacts are visible JPEG compression works on small square blocks across the image. When your source image contains a hard transition between two very different colors within a single block, the JPEG step approximates that transition imperfectly. The result is ringing, a halo of lighter or darker tones around the edge. In photographs this is invisible, because edges are never perfectly sharp and real-world photos carry micro-variation that the approximation matches well. In screenshots, logos, UI elements with hard borders, text, or adjacent flat color areas, the ringing shows clearly at any quality setting, because those elements are exactly the kind of signal the algorithm handles poorly. The advice is straightforward: do not convert screenshots, logos, or text-heavy images to JPG. Keep them as PNG, or use WebP for a smaller file that avoids the problem entirely. ### How conversion paths affect your file Where the conversion happens depends on the number of files. For a single image, your file is converted locally on your device, and no image data travels over the network. You can verify this by watching your browser's network activity during a single conversion and seeing no outbound request carrying your image. For two or more files, RoundCut sends them to our server, which converts them, packages the result, and returns a download link. That link and the converted files are deleted within about 2 hours. The single-image path runs entirely on your device, while the batch path trades that for the convenience of converting many files in one step. For images with sensitive content, such as a personal document or a photo with location data, converting one at a time keeps everything local. --- # Convert PNG to WebP Source: https://roundcut.app/png-to-webp/ > Convert PNG to WebP with transparency preserved and a smaller file size. RoundCut converts PNG to WebP fast, with transparency preserved. A single image is converted on your device with no upload. Convert several at once and they go to our server to be processed together into one download, and that result is deleted within about 2 hours. Transparent areas in the PNG stay transparent in the WebP output, because WebP carries a full transparency support, which makes it the right choice when you need a smaller file without losing the transparent cut-out that JPG conversion would destroy. Per Google's own WebP documentation, the exact-copy WebP mode is about 26 percent smaller than an equivalent PNG, and compressed WebP with an transparency support is roughly three times smaller. The output is compressed WebP at a near-exact quality setting, meaning some fine bit-level detail is traded for a smaller file, but the visual result is indistinguishable from the source PNG at normal viewing sizes. For precise archival work, keep the original PNG. ## What it does - **Transparency preserved** WebP has a full transparency support, so transparent PNG pixels stay transparent - **One image stays local** converting a single PNG runs on your device, with no file data uploaded - **No extra download** WebP encoding uses the platform image engine, so no codec module is fetched - **Smaller output** compressed WebP with transparency is typically 3 times smaller than equivalent PNG per Google's benchmark - **compressed output** this tool produces compressed WebP, where bit-level detail is traded for smaller size - **Metadata removed** EXIF, IPTC, and XMP are stripped, and ICC profile handling varies by browser - **No animated output** an animated PNG is processed as a single still, the first frame - **Single file per run** one PNG at a time, with batch conversion planned for a later release ## How to use it 1. **Drop or pick your PNG** — Drag a PNG file onto the upload area or click to open a file picker. Any transparent areas in the PNG will be preserved in the output. Converting one image runs on your device, while dropping several at once sends them to our server to process together. 2. **Wait for the automatic conversion** — The browser encodes the WebP the moment the file loads. For most images at screen resolution, the result appears in well under a second with no button to press. 3. **Check the output size** — The Download button shows the output file size. Transparent areas remain transparent in the WebP, and the file should be noticeably smaller than the source PNG. 4. **Download the WebP** — Click Download to save the file. The original filename is kept with the extension changed to .webp. To convert another PNG, drop it onto the page, and each file runs independently. ## How to convert PNG to WebP Drop a PNG onto the upload zone, or click to pick a file from your device. The conversion starts the moment the file lands, with no Convert button to press. The result appears in well under a second for typical images. When it is ready, the Download button saves the WebP using the original filename with the extension changed. Converting one image runs on your device, while converting several at once sends them to our server. The same code path works on desktop and mobile browsers without any app install. To convert another image, drop the next PNG. An animated PNG is processed as a single still frame from the first image in the sequence. ## Transparent pixels pass through to WebP WebP supports full transparency, which means every transparent area in your PNG stays exactly as transparent in the WebP output. Nothing is filled with white, no halo appears around cut-outs, and no editor touch-up is needed afterward. This is the defining reason to choose WebP over JPG when converting from a PNG with transparency, since JPG cannot carry transparency and must replace transparent areas with a solid color. Logos, icons, product cut-outs, UI mockups, screenshots with rounded corners, and any image that relies on a see-through background all transfer intact. There is no toggle to enable transparency preservation. It is automatic because WebP is designed to carry full transparency. ## Reasons to switch from PNG to WebP WebP was designed as a successor to PNG and JPG for web delivery, and it achieves better compression by combining two distinct compression modes. The exact-copy mode is better than PNG's compression algorithm by about 26 percent on typical images, per Google's benchmark. The compressed mode with transparency is roughly three times smaller than a comparable PNG, making it the format of choice when you need both a smaller file and a see-through background. Web performance tools notice this directly. Largest Contentful Paint, one of Google's Core Web Vitals and a known ranking signal, improves when your largest image loads faster. Switching large PNG assets to WebP is one of the most direct ways to move that number. All major browsers have supported WebP natively since 2020, so there is no compatibility concern for current audiences. ## Quality settings and the size trade-off WebP gives you a quality dial from 0 to 100, where higher values preserve more detail at the cost of a larger file. RoundCut encodes at a fixed setting near the top of that range, chosen to balance visual fidelity against file size for typical web assets: photographs, UI graphics, and icons. At this setting, size reductions on photographs typically land between 50 and 70 percent compared to the source PNG, on top of the structural advantage WebP already has over PNG. For graphics with large flat-color areas, the reduction is even larger. If you need a perfect byte-for-byte copy that can be decoded back to identical values, keep your original PNG and treat the WebP as the delivery copy. The source PNG is never changed by the conversion, only the exported WebP changes. ## Where your PNG is processed The path depends on how many files you convert. A single PNG is converted on your device using the browser's built-in image engine, with nothing uploaded. You can verify this by watching your network panel while you convert one image and confirming zero outbound requests carry it. Convert several PNGs in one go and they are sent to our server, which converts them together and returns one download for you. That result is deleted from our server within about 2 hours, and you can also delete it yourself the moment you have saved it. So a single conversion stays entirely on your device, and a batch is handled on our server only long enough to build your download. ## When the original PNG is still the right choice WebP is widely supported but not universally accepted in every context. Some design tools, print workflows, and internal content-management systems still require PNG. Some operating systems handle PNG drag-and-drop natively where WebP needs a plugin. For working files you will open and edit repeatedly in a desktop application, keeping the without quality loss PNG as the master prevents any accumulation of decode-and-re-encode cycles. Use WebP as the export format for web delivery and keep the PNG as the source. If you need to go the other direction and turn a WebP back into PNG, the webp-to-png tool handles that conversion. ## Questions **Will transparency be preserved when converting PNG to WebP?** Yes. WebP supports a full transparency support, so transparent areas in your PNG stay transparent in the WebP output. This is the main reason to choose WebP over JPG when starting from a PNG, since JPG has no transparency and must flatten transparency to a solid color, while WebP keeps the cut-out exactly as it was. Logos, icons, product images, and UI assets with transparent backgrounds all pass through the conversion intact with no extra steps. **Why convert PNG to WebP?** Smaller files without losing transparency or visual quality. Per Google's WebP documentation, exact-copy WebP is about 26 percent smaller than equivalent PNG, and compressed WebP with transparency can be roughly three times smaller. Smaller images improve page load speed and directly help Largest Contentful Paint, a Core Web Vitals metric and a known Google ranking signal. PageSpeed Insights flags PNG as a next-gen format opportunity in its image audit. Browser support is universal across Chrome, Firefox, Safari, and Edge. **Does converting PNG to WebP lose quality?** At the near-exact setting used by this tool, photos and graphics are visually indistinguishable from the source PNG at normal viewing sizes. WebP at this setting is technically compressed, meaning some bit-level precision is traded for a smaller file. For typical web use, including logos, icons, and photographs, that difference is not visible. For precision or archival work, keep the original PNG and treat the WebP as the delivery export. The source PNG is not modified by the conversion, it remains unchanged on your device. **How much smaller is WebP than PNG?** Per Google's official WebP documentation, exact-copy WebP is around 26 percent smaller than an equivalent PNG, and compressed WebP with an transparency support is roughly three times smaller than a comparable PNG. The exact ratio depends on image content, where graphics with large transparent areas and solid-color regions see the biggest gains, while highly detailed photographs see smaller but still meaningful reductions. Either way, the WebP is smaller and the transparency is preserved. **Is it safe to convert PNG to WebP here?** It depends on whether you convert one image or several. A single PNG is converted on your device with no upload, which you can confirm by watching your network panel during the conversion and seeing no outbound image request. When you convert several at once, they are sent to our server to be processed together, and that result is deleted within about 2 hours, with an option to delete it yourself as soon as you download it. Either way, nothing about your image is kept on any server beyond building your download. **Does WebP work in all browsers?** Yes. WebP is supported natively in Chrome, Firefox, Safari, and Edge, covering the substantial majority of browsers in active use. Support has been universal since Safari 14 in 2020. If you are building for an audience that might include very old browsers or embedded webviews, PNG remains the safest choice. For any audience using current browsers from the past five years, WebP works reliably. You can verify per-browser support on the caniuse.com WebP entry. ## The details ### How WebP achieves smaller sizes than PNG WebP uses two distinct compression modes. The exact-copy mode combines spatial and color prediction with an entropy coding stage that is more sophisticated than PNG's DEFLATE, achieving compression ratios roughly 26 percent better on typical images according to Google's published benchmark. The size-reducing mode applies a block-based transform similar to video compression, designed to discard perceptually irrelevant information while preserving what the eye actually sees. For images with an transparency support, WebP uses a separate no-loss sub-compression for the transparency data while applying size-reducing compression to the RGB data, which is why compressed WebP with transparency can be roughly three times smaller than a PNG at equivalent visual quality. PNG's DEFLATE is without quality loss-only and cannot take advantage of the perceptual trade-off that the size-reducing mode exploits. The format difference explains the gap. ### Transparency in detail Both PNG and WebP support 8-bit transparency transparency, meaning each detail can hold an opacity value from 0 (fully transparent) to 255 (fully opaque). When re-encoding a PNG to WebP, the engine reads every color value and passes the transparency values into the WebP compression step. Those transparency values are preserved with no-loss compression, independent of how the color data is encoded. The result is that fully transparent areas, semi-transparent areas, and fully opaque areas all map to the exact same state in the WebP. For a logo with soft drop-shadows or anti-aliased text on a transparent background, the edge smoothness survives intact. JPG has no transparency field in its format specification and must fill transparent areas with a background color before encoding. ### Core Web Vitals and image format choice Largest Contentful Paint measures how long it takes for the largest visible element on the page to appear in the viewport. For most content pages, that element is the hero image, and Google uses LCP as a ranking signal in its Core Web Vitals assessment. PNG images are a consistent source of LCP failures. A photographic PNG at 4K resolution can weigh several megabytes, while the equivalent WebP at near-exact quality is typically a fraction of that. PageSpeed Insights flags this specifically in its serve-images-in-next-gen-formats audit item, listing PNG as the target format to replace. Converting PNGs that appear in critical render paths to WebP is one of the highest-leverage single changes for improving measured page performance. The browser support timeline makes this safe, with Chrome adding WebP in 2011, Firefox in 2019, Safari in 2020, and Edge from its Chromium rebuild. ### Compressed output and the near-exact setting WebP encoding offers a quality parameter from 0 to 100, where higher values preserve more detail at the cost of larger files. This tool encodes at a fixed near-exact setting tuned to balance visual fidelity against file size for the typical web asset categories of photographs, UI graphics, and icons. At this setting, the output is perceptually indistinguishable from the source PNG at normal screen viewing distances. Technically, some bit-level precision is lost relative to a true without quality loss encode, which means a byte-level comparison between the decoded WebP pixels and the original PNG pixels will show small numerical differences. These differences are below the threshold of human perception for photographic content. For medical imaging, satellite photography, or archival preservation, fidelity without any loss is a hard requirement. In those cases, keep the PNG as the working copy and use WebP only as a delivery export. ### Metadata behavior The pipeline used for PNG-to-WebP conversion strips EXIF, IPTC, and XMP metadata from the output. This matches the behavior observed across all three major browser engines. ICC color profiles are handled inconsistently, where Chrome and Safari preserve the sRGB ICC profile tag in WebP output and Firefox strips all metadata including the ICC profile. The net result is that the converted WebP is sRGB-safe across browsers, but any wide-gamut tagging such as Display-P3, Adobe RGB, or ProPhoto RGB present in the source PNG will not survive in Firefox. For color-critical professional workflows, this inconsistency matters. If the final output needs color profile fidelity, convert with a tool that explicitly writes ICC data. Or apply the profile tag as a post-processing step using a dedicated metadata editor. ### When to keep PNG and when WebP is sufficient The practical decision tree is short. If the image's final destination is a web page or web application, and the rendering environment is any browser from 2020 onward, WebP is the right export format. If the image needs to open in a design application like Figma, Sketch, or Affinity Designer, verify WebP support in the specific version you use. Support varies across versions. If the image will be used in a print workflow, PNG or TIFF is preferred since most print RIPs do not handle WebP. If the image will be sent in email, PNG is safer since email clients are notoriously inconsistent with modern formats. If the image will be used as a working file that gets edited and re-saved multiple times, keep the PNG as the master. WebP at near-exact quality is an excellent delivery format, and PNG is the better archival and editing format. The ideal workflow keeps the PNG as the original and exports to WebP for web delivery. --- # Convert WebP to AVIF, Smaller Files With Alpha Kept Source: https://roundcut.app/webp-to-avif/ > Convert WebP to AVIF and get photographs that run 20 to 30 percent smaller, with transparency kept. Convert a WebP to AVIF in a couple of clicks. Because AVIF is the hardest format to encode, our server does the heavy lifting to give you the best result, so converting to AVIF may send the image to our server for processing. When several files are converted at once, our server always handles the batch and the download link is deleted within about 2 hours. Both formats compress with quality loss, so this is a second compression pass: color detail already compressed by the WebP encoder is compressed again. For photographs the result is visually near-exact at quality 85, around 42 dB PSNR, and the output is typically 20 to 30 percent smaller than the source WebP. If your WebP has a transparent background, the transparency survives intact. Encode speed is the honest caveat. A 2-megapixel photo takes under a second, an 8-megapixel photo a few seconds, and larger images take longer. The first conversion of a session can also pay a brief one-time warm-up. ## What it does - Typically 20 to 30 percent smaller than WebP on photographs at comparable perceptual quality, with the largest savings on natural images. - Alpha channel preserved, both WebP and AVIF support transparency and the conversion carries it through without flattening. - A second lossy pass, output quality is bounded by the source WebP, so a truly lossless AVIF is not possible from a lossy WebP. - AVIF encoding uses our server for the best result and the download is deleted within about 2 hours, with a quiet in-browser fallback if the server cannot be reached. - Encoded at quality 85, around 42 dB PSNR on photographs, near-indistinguishable from the source at normal display sizes. - Warm-up on the first encode, the AVIF encoder loads once per session and later encodes in the same tab skip that step. - Honest encode speed, 8 megapixels takes roughly 3 seconds on Chrome and Firefox runs about four times slower. - One WebP in, one AVIF out, single file per conversion in this version. ## How to use it 1. **Drop your WebP** — Drag the WebP onto the upload area, or click it to open a file picker and choose one from your device. 2. **Let it convert** — AVIF is the heaviest format to encode, so the image is sent to our server for the best result. The download is removed within about 2 hours. 3. **Review the sizes** — When the AVIF is ready, the stats line shows the WebP source size and the AVIF output size. A 20 to 30 percent drop on photographs is typical. 4. **Download the AVIF** — Click Download to save the file to your device with the same base name and a new .avif extension. ## How to convert WebP to AVIF Drop your WebP onto the upload area or click it to pick a file. The conversion starts the moment the file lands, with no separate convert button. AVIF is the heaviest format to encode, so a single image is sent to our server to produce the best result, and the AVIF comes back ready to download. A small image converts in well under a second, a 2-megapixel photo in about a second, and an 8-megapixel photo in a few seconds. When the AVIF is ready, click Download to save it with the same base name and a new .avif extension. The stats line reports both file sizes so the reduction is visible, and the download link is deleted within about 2 hours. ## How much smaller is AVIF than WebP? AVIF is built on a modern codec and reaches files roughly 20 to 30 percent smaller than WebP at the same perceptual quality, based on SSIM and DSSIM comparisons across photo test sets. On simple graphics with flat color or hard edges the gap narrows, and WebP can pull even or ahead. The win is largest on photographs, exactly the assets that dominate page weight on most modern sites. A source WebP weighing 32 kilobytes might become 22 to 26 kilobytes as AVIF at equivalent visual quality. For heavily compressed WebP sources the saving narrows further. AVIF browser support now passes 93 percent globally, so the fallback case for modern audiences has shrunk to a small minority. ## Does WebP to AVIF preserve transparency? Yes. WebP supports transparency and so does AVIF, and the conversion carries it through unchanged. A WebP with a transparent background converts to an AVIF with the same transparent background, no white fill, no color shift, no edge halo. This is the opposite of converting to JPG, where transparency is destroyed because JPG has none. If your WebP holds a cut-out logo, a UI sprite, or a sticker on a transparent canvas, AVIF is a safe destination and you do not need to flatten the image first. The transparent layer is encoded alongside the color data at quality 85 and arrives in the output with soft edges intact. ## Quality and encode speed Both formats compress with quality loss, and this pair runs at quality 85. At that setting, photographs measure around 42 dB PSNR, visually near-exact for natural images. Graphics with hard edges or flat color can show mild artifacts. Since the source is already a compressed WebP, this is a second compression pass: information the WebP encoder already dropped cannot be recovered. Speed is the trade-off for AVIF's compression efficiency. On Chrome desktop, a 2-megapixel photo finishes in under a second, an 8-megapixel photo in roughly 3 seconds, and a 24-megapixel shot in 10 to 30 seconds. Firefox runs about four times slower, and a phone is 3 to 5 times slower than desktop Chrome. The first encode also pays a one-time warm-up. ## AVIF browser support for delivery AVIF now reaches over 93 percent of global browsers: Chrome 85 and later, Firefox 93 and later, Safari 16.4 and later, and Edge 121 and later all display it natively. Internet Explorer, Opera Mini, and pre-2022 Safari do not. The practical upshot for delivery is that an HTML picture element serving AVIF with a WebP fallback covers essentially all modern traffic. For sites where compatibility with the remaining few percent matters, keep the WebP fallback in place. For sites confident in a modern-browser audience, AVIF alone is viable. The bandwidth saving makes AVIF first with a WebP fallback the dominant pattern among performance-focused developers in 2026. ## Where your file is processed AVIF is the most demanding format to encode, so this pair uses our server to produce the best result, which means converting to AVIF can send your WebP to our server. The work happens, the AVIF comes back, and the download link is deleted within about 2 hours. When you convert several files in one go, our server always handles the batch and zips the results for you, again with the same roughly 2-hour deletion window. If the server is ever unreachable, the page quietly encodes a single image locally instead, keeping that one file on your device. We hold no permanent copy of your image beyond that short processing window. The honest summary is simple: a single AVIF may be made on our server for quality, a batch always is, and nothing of yours is kept around afterward. ## Questions **Is AVIF better than WebP?** For photographs delivered to a modern browser audience, AVIF wins on compression: typically 20 to 30 percent smaller files at the same perceptual quality, based on DSSIM measurements across photo test sets. On graphics with flat color and hard edges, WebP is competitive and can occasionally produce a smaller file. Both support alpha and animation. If your audience uses modern browsers, currently around 93 percent of global users, AVIF is the better default. Otherwise, keep WebP as a fallback. **Will I lose quality when converting WebP to AVIF?** Yes, some generation loss is unavoidable. Both formats compress with quality loss, so re-encoding a compressed WebP at AVIF quality 85 adds a second compression pass. For photographs the result is visually near-exact, measuring around 42 dB PSNR. For graphics with hard edges or flat color, expect mild artifacts. A truly exact AVIF is not possible from a compressed WebP source, since the information the WebP encoder discarded is already gone before this conversion starts. **Does AVIF support transparency like WebP?** Yes. Both WebP and AVIF support transparency, and the conversion carries it through exactly, with no white fill, no color shift, and no edge halo. This differs from converting to JPG, where transparency is permanently destroyed because JPG has none. If your WebP has a transparent background, your AVIF will too, with the same soft edges and partially transparent areas intact. No manual flattening is needed before you convert. **How long does WebP to AVIF conversion take?** The first encode of a session loads the encoder in about a second. After that, a 2-megapixel image finishes in under a second on Chrome, 8 megapixels takes roughly 3 seconds, and 24 megapixels can take 10 to 30 seconds. Firefox runs about four times slower on AVIF than Chrome, so a large image on Firefox can stall for half a minute. A phone is 3 to 5 times slower than desktop. For large files or Firefox users, Chrome on a desktop is the fastest path. **Why is AVIF encoding slower than WebP?** AVIF is built on a modern codec that applies far more complex prediction and entropy coding than WebP's older algorithm. The newer format was designed to maximize compression at the cost of encode complexity, accepting slow encode in exchange for best-in-class output. Decoding AVIF is fast in all modern browsers because hardware decode is standard. But encoding is heavy work, and the computation for a large photograph is genuinely intensive, which is why the encode is not instant. **When should I convert WebP to AVIF?** When you control the delivery environment end to end and your audience uses modern browsers, upgrading from WebP to AVIF makes sense for photographs and photo-like content where the 20 to 30 percent size saving translates into faster LCP scores. It is less compelling for small icons or simple graphics where WebP is already compact, for heavy-traffic pages where the encode time would stall a pipeline, or when the destination system requires WebP specifically. ## The details ### Why AVIF beats WebP on photographs AVIF's compression advantage over WebP comes from the underlying codec. WebP uses the an engine built for video back in 2010. The codec behind AVIF was released in 2018 after years of research aimed at the limits of those older methods. It compresses color detail more aggressively, divides the image into smaller blocks, and packs the result more tightly. The practical result, confirmed across large photo test sets, is that AVIF is typically 20 to 30 percent smaller than WebP at the same perceptual quality on photographs. On computer-generated graphics with flat color and hard edges the gap narrows, because WebP's simpler algorithm handles those patterns efficiently. For any page whose heaviest assets are photographs, the AVIF upgrade has measurable bandwidth impact. ### The second lossy pass in practice When you convert a WebP to AVIF, both the source and the output use compression. The WebP encoder already decided which color information to discard when the WebP was first created, and those decisions are permanent: converting to AVIF cannot recover any of it. The AVIF encoder then makes its own decisions about the color values it receives, at quality 85. For photographs, the combined effect of two compression passes is still visually near-exact at typical display sizes, measuring around 42 dB PSNR on real photo content. For graphics already degraded by their WebP encoding, the second pass can compound visible artifacts. The practical rule: if the source WebP looks clean at the size you plan to display, the AVIF will too. If the WebP already shows compression noise, examine the AVIF carefully before shipping it. ### How the alpha channel flows through Both WebP and AVIF encode transparency as a separate layer alongside the color data. The conversion reads the WebP's transparency mask, composites it at full opacity, then hands it to the AVIF encoder, which writes its own transparency track using intra-frame coding at quality 85. Soft transparency gradients, feathered edges, and partially transparent areas all survive this round trip. The AVIF transparency layer is itself compressed at quality 85, which can introduce barely perceptible fringing at hard edges under extreme zoom. At normal viewing sizes the difference from the source is not visible. For precise work on small icons where the transparent edge must be exact, keep the WebP source and verify the output at full zoom before you ship. ### Encode-time expectations across browsers and devices The AVIF encoder warms up once per browser session, adding about a second to the first conversion. After that, Chromium gives the best throughput: roughly 40 milliseconds for a 0.12-megapixel thumbnail, 250 milliseconds for a 1-megapixel photo, 2.8 seconds for a 4K photo, and 25 seconds for a full 48-megapixel image in the worst case observed. Firefox is the notable outlier, running the same encoder about four times slower, which puts the 4K case near 31 seconds and large files past two minutes. WebKit sits between the two, closer to Chromium. Mobile hardware runs 3 to 5 times slower than desktop across all engines. For routine file-by-file conversion, Chrome on a laptop or desktop is the practical tool. For Firefox users with large images, the honest advice is to switch browsers or keep WebP. ### Core Web Vitals and the AVIF upgrade Pages that serve photographs pay a bandwidth cost for every image loaded. If those images are currently WebP, converting them to AVIF cuts the per-image transfer by roughly 20 to 30 percent on photographs. On a page with a 200-kilobyte WebP hero image, AVIF brings that to roughly 140 to 160 kilobytes. On a product grid with twelve 30-kilobyte WebP thumbnails, AVIF saves roughly 70 to 90 kilobytes of total page weight. These savings directly affect Largest Contentful Paint when the LCP element is an image. At about 94 percent global browser support for AVIF, a picture element serving AVIF first with a WebP fallback covers essentially all traffic. The markup cost is paid once per image component, and the bandwidth saving repeats across every page load afterward. ### Why AVIF conversion leans on a server AVIF is built on a modern codec, the same family used for modern video, and it is deliberately heavy to encode in exchange for its tight files. Doing that work well, especially on large photographs, is far faster and more reliable on a real server than in a phone browser, so this pair sends the image to our server to encode the best possible AVIF. The file is processed and the result is handed back, and the download is removed within about 2 hours, with no long-term storage. Converting several images at once always runs on the server, which assembles the finished set into a single download for you, deleted on the same short timer. If our server cannot be reached, a single conversion is encoded locally instead, that one file staying on your device, accepting a slower encode for the convenience. The trade is plain: server encoding buys you quality and speed on the format that needs it most, and your file is never retained past the brief window it takes to convert. --- # Convert WebP to JPG, a File That Opens Everywhere Source: https://roundcut.app/webp-to-jpg/ > Turn a WebP into a standard JPG for email, Instagram, and apps that still want JPEG. Works right in the tab, no install needed. Convert WebP to JPG straight from the browser tab. Drop a WebP and your browser writes out a compatible JPG in about a second, with a single file converting on your own machine and nothing uploaded. JPG holds no transparency, so any clear areas in your WebP get filled with white. That is a built-in property of JPEG, not a quirk of this tool. The output loses some detail in encoding, just as the original WebP did, and at the default quality the difference on photos is hard to spot. That JPG then opens everywhere: Instagram, WhatsApp, Outlook, Windows File Explorer, print services, and older copies of Photoshop. WebP runs great in modern browsers, but a long list of platforms still take only JPEG. Need to convert several at once? Drop them together and they go to our server to process as a batch, where the download link is removed within about 2 hours. ## What it does - Native JPEG output. Your browser's own JPEG writer does the work, with no plugin or extra download. - One file converts locally on your machine. Convert several at once and they go to our server, where the download link is removed within about 2 hours. - Transparency filled with white. Clear or partly clear WebP areas become white in the JPG, because JPEG holds no transparency. - Under a second. A typical phone photo finishes the full round trip in about 16 milliseconds on Chromium in our tests. - Opens anywhere. JPG works on every platform, every operating system, every app, and every upload form with no codec to install. - Metadata dropped. The output JPG carries no EXIF, IPTC, XMP, or GPS data, only the visible detail. - Auto-renamed download. The file keeps its original name and changes to the .jpg extension for you. - Zero added page weight. JPEG output uses the browser codec and ships no extra bytes. ## How to use it 1. **Drop or pick your WebP** — Click the upload area or drag a WebP onto it from your desktop. A single file converts the moment it lands, right on your machine, with nothing sent over the network. 2. **Let the JPG encode** — The browser's JPEG writer runs a single file right on your machine. A typical phone photo finishes in under a second, with nothing to upload. 3. **Read the output sizes** — A stats line shows the original WebP size and the resulting JPG size side by side, so you can confirm the format change before you download anything. 4. **Download the JPG** — Click Download to save the result. The file keeps its original name and switches to the .jpg extension for you, so nothing needs renaming. ## Why convert WebP to JPG You saved a picture from a website and it arrived as a .webp file. Instagram will not take it for an upload. Your email client refuses to attach it. A job portal, a school system, or an older content manager rejects it with a vague error. WebP is Google's efficient modern format, and every current browser shows it fine, but a long tail of apps, platforms, and systems still accept only JPEG. Convert the file here and you get a plain JPG that uploads anywhere a normal photo would, with no app to install and no extension to add. It is the everyday fix for an everyday compatibility headache. ## How to convert WebP to JPG in three steps Drag the WebP onto the upload area, or click to pick one from your device. The conversion starts the second the file lands, with no Convert button, no upload wait, and no queue. When the JPG is ready, a stats line shows the input size and output size, and the Download button drops the file onto your device. The whole thing usually takes under a second for a phone photo. The page stays put between conversions, so you can drop the next file right on top of the last one without reloading. It is built to feel instant for a single file rather than to manage a long batch. ## What happens to transparency and quality JPG is a format with no transparency channel. Converting WebP to JPG permanently sheds some colour detail, as every encode does. At the default quality the difference is hard to notice on photos, but this is not an exact conversion and no tool can make it one. Clear areas in the source WebP get filled with solid white, because the JPEG format simply cannot store transparency. If you need to keep that transparency, convert to PNG instead. If you need a background colour other than white, paint it onto the WebP in an editor before you convert here. Any tool that claims an exact WebP to JPG path without loss is technically incorrect, and worth a second look. ## JPG versus WebP, which one to keep WebP and JPG both compress photographs, but compatibility decides which one to keep. Stay with WebP when the image serves a web page you run and lighter files matter for visitors on modern browsers. Switch to JPG when the destination is outside your control, like an Instagram upload, a WhatsApp attachment, a print lab, a Windows PC without extra codecs, an older photo editor, or a marketplace upload form. If the image reaches your audience through your own website, WebP earns its keep. The moment it leaves and heads somewhere else, JPG is the safer bet. Reach matters more than efficiency once the file goes beyond your own domain. ## Where your file is handled Converting one file runs on the browser's own JPEG writer, right on your own machine, with nothing uploaded. You can confirm this by watching the Network panel in the developer tools while you convert a single file. You will see no outbound image request in the trace. When you convert several at once, the files go to our server to do the work together, and the download link is removed within about 2 hours. Either way the result is yours to keep, and nothing is logged or studied. ## Animated WebP and what happens to it The converter keeps only the first frame of an animated WebP. Every later animation frame is dropped, and the output is a still JPEG of that opening frame. JPG itself has no animation support to fall back on. If you need to work with the full sequence, reach for a dedicated animated-image tool or a video converter instead. This limit applies across the whole in-browser convert pipeline, since it comes from how the browser decodes WebP animation rather than from a choice we made. For a single still photo, none of this matters and the conversion behaves exactly as you expect. ## Questions **Why convert WebP to JPG?** WebP is not accepted everywhere. Instagram, WhatsApp, Windows File Explorer without the extra codec, older Photoshop versions, and most print services still want JPEG. JPG works on every device, every operating system, every social platform, and every editor. Converting trades WebP's slimmer file size for guaranteed compatibility wherever you actually need to open, attach, or send the picture, which is usually the whole point of the conversion in the first place. **Will transparency be preserved in JPG?** No. JPEG has no transparency, so transparent or partly transparent areas in your WebP become solid white in the output. No setting changes this, because it is a constraint of the JPEG format itself, not of this tool. If you need to keep transparency, convert to PNG instead. If you need a different background colour, paint that colour onto the WebP in an image editor before you convert it here, and the JPG will show your chosen background. **Will the conversion reduce image quality?** Yes. JPEG discards some colour detail on every encode, and so does WebP, so every encode of either format permanently drops some information. At the default quality used here, the visible difference on photographs is hard to notice. On images with sharp edges or fine text, faint JPEG artifacts may appear. Any tool that promises an exact WebP to JPG conversion without loss is incorrect. If you need precise preservation, convert to PNG instead of JPG. **Is my image uploaded to a server?** It depends on how many. Converting one file runs entirely in the tab, in the browser's own image engine, with no outbound image request. You can confirm this by watching the Network panel in the developer tools. Converting several at once sends them to our server so they process together, and the download link is removed within about 2 hours. There is no telemetry on your file contents and no long-term storage either way. **What is a WebP file and why do I have one?** WebP is an image format Google built for smaller files on the web. You most likely got one by right-clicking and saving a picture from a modern site, since browsers save whatever the server sends and most servers now send WebP to browsers that support it. WebP is excellent for web delivery, but a large number of apps, upload forms, and email clients still take only JPEG, which is exactly why converting it is such a routine task. **Can I convert several WebP files to JPG at once?** Yes. Drop two or more WebP files and they convert together, then come back as one download. To do this the files go to our server, which packs the JPG results into one archive and gives you a single link that is removed within about 2 hours. A single WebP still converts right in the tab with nothing uploaded. ## The details ### Why so many apps still reject WebP in 2026 Even though WebP is more than a decade old, a real slice of consumer software still will not accept it for upload or display. The reason is not technical capability, since the codec is widely available, but product inertia and format standardisation. Instagram and the wider Meta platforms standardised on JPEG for upload because it is the most portable format for a global user base, and changing upload validation means re-testing across hundreds of device setups. Microsoft's photo viewer on Windows 10 needs a separate Store install to handle WebP. Most enterprise content managers built before 2018 validate uploads against a fixed MIME list that predates WebP entirely. Print providers, stock agencies, and government portals often bake format requirements into legal or procurement specs. These are genuine barriers that demand a compatible format, and JPEG remains the universal answer to them. ### The transparency problem: why white, and how to change it JPEG's compression works on colour channels pulled from red, green, and blue values. There is no fourth channel for transparency. When the browser writes a JPEG from a WebP that carries clear areas, it has to assign a real colour to each transparent spot. The default fill is white. That is the browser default, not a setting this tool exposes. If your WebP has a transparent background and you need a particular colour in the JPG, say dark grey for a logo on a dark page, open the WebP in any editor that supports layers, drop a background layer in your target colour beneath the image, flatten the stack, and then run that flattened file through this converter. The result is a JPG carrying your chosen colour exactly where the transparency used to be. ### Quality, PSNR, and what visually negligible means PSNR, Peak Signal-to-Noise Ratio, is the standard engineering measure of how far a re-encoded image drifts from its source, expressed in decibels. A higher figure means a closer match. In our end-to-end tests, the JPEG writer used here produced 43.66 dB PSNR on a photograph at the default quality. For context, 36 dB is the rough threshold below which differences become clearly visible to most viewers, while 40 dB and up is usually called perceptually transparent for photographic content. At 43.66 dB, the conversion is, for any practical web or print purpose, visually identical to the source. The quality is fixed in this version with no slider. If you need a precise quality level for a particular workflow, a more configurable tool is the better fit for that job. ### File size after conversion: WebP to JPG is usually larger WebP compresses photographs more efficiently than JPEG does. When you convert in the reverse direction, from WebP to JPG, you move to a less efficient format, so the output usually ends up larger than the input. In our measurements, a WebP photograph at a standard phone resolution produced a JPEG that was a touch smaller in that one case because JPEG happened to encode that image well. But on larger photos the inflation is consistent. A picture that arrives as a compact WebP can leave as a JPEG roughly 50 percent heavier at comparable visual quality. This is not a bug or a sign the converter is misbehaving. It is the expected price of moving from a more efficient format to a less efficient one, and that growth is simply the cost of universal compatibility. ### Metadata: EXIF, GPS, and what gets stripped WebP files can carry EXIF metadata, including GPS coordinates, camera model, shutter speed, aperture, and orientation. JPEG files can carry EXIF too. The conversion pipeline here, though, strips all metadata from the output. The resulting JPEG holds only visual detail, with no EXIF, no IPTC copyright fields, no XMP edit history, and no ICC profile in Firefox or WebKit, while Chrome and Edge do preserve ICC profiles. This happens at the browser drawing level and cannot be overridden. For most web publishing, stripped metadata is the preferable outcome: it trims a few bytes, removes GPS coordinates that might pin down where a photo was taken, and avoids leaking camera or editing details. For archival or legal work where metadata has to survive, use a dedicated EXIF-preserving editor instead of this converter. ### How long does WebP to JPG conversion take? In our end-to-end tests across Chromium, Firefox, and WebKit, the JPEG encode step for a standard phone photo took 16 milliseconds on Chromium, 12 on Firefox, and 17 on WebKit. For a large photo at a high resolution, the encode step added roughly 1.5 seconds on Chromium. The total time you actually feel also folds in the browser reading the file from disk, the WebP decode step, and the surface draw before the encode. For phone photos under four megapixels, the round trip stays under a second. For big DSLR photos above eight megapixels, expect 2 to 5 seconds on a desktop browser. Mobile devices run roughly three to five times slower than desktop for the encode alone, which tracks with the gap in mobile processor performance. --- # Convert WebP to PNG Source: https://roundcut.app/webp-to-png/ > Convert WebP to PNG so Photoshop, Word, and older apps can open the file. Transparency kept. RoundCut converts WebP to PNG so the file opens everywhere desktop software expects. Convert one image and the browser re-encodes it as PNG on your own machine in under a second. Convert several at once and they go to our server to be processed together into one download, which is then removed within about 2 hours. PNG is the format that desktop applications actually open, since Photoshop in older versions, Word, PowerPoint, Preview, and most legacy tools on Windows and macOS still refuse WebP files. Transparency is preserved exactly as it was in the source, because both WebP and PNG carry an 8-bit transparency channel, so a logo on a clear background converts to a PNG with the same clear background, with no white fill and no edge artifacts. The output PNG is always larger than the source WebP. This is the expected cost of switching to a format that opens everywhere. No new quality loss is introduced by the conversion itself. ## What it does - **Transparency preserved** both WebP and PNG carry the same 8-bit transparency data, so clear and semi-clear areas pass through unchanged - **One image converts on your machine** converting a single WebP runs in the browser tab, with no image data sent over the network - **No extra download** PNG encoding uses the platform image engine, so no codec module is fetched - **No new quality loss** PNG stores detail exactly, so the existing WebP appearance is kept without any further degradation - **File grows in size** PNG is exact and WebP is more compressed, so the PNG is always larger - **Metadata removed** EXIF, IPTC, and XMP are stripped on all browsers, and ICC handling varies by browser - **First frame only** a multi-frame animated WebP produces a single still PNG of frame one - **Single file per run** one WebP at a time, with batch conversion planned for a later release ## How to use it 1. **Drop or pick your WebP** — Drag a WebP file onto the upload area or click to open a file picker. Transparency in the WebP will be preserved in the output. Converting one file runs on your machine, while dropping several at once sends them to our server to process together. 2. **Wait for the automatic conversion** — The browser writes the PNG the moment the file loads. Most images at screen resolution finish in under a second with no button to press. 3. **Note the output size** — The PNG will be noticeably larger than the source WebP. A large photographic WebP can grow to several times its original size as PNG. This is expected and not an error. 4. **Download the PNG** — Click Download to save the file. The original filename is kept with the extension changed to .png. Transparent areas remain transparent. To convert another file, drop the next WebP. ## How to convert WebP to PNG Drop a WebP file onto the upload zone, or click to pick a file from your device. The conversion starts the moment the file lands, with no Convert button to press. The result appears in under a second for most images. When it is ready, the Download button saves the PNG using the original filename with the extension changed to .png. To convert another image, drop the next file. Each conversion runs from scratch with no queue. The code works the same on desktop and mobile browsers. An animated WebP is processed as a single still frame, so only the first frame appears in the output PNG. ## Transparent backgrounds stay transparent Both WebP and PNG support full transparency, meaning each area of your image can carry any opacity value from fully clear to fully solid. When the browser reads the WebP and re-encodes it as PNG, the transparency values are transferred intact. A logo on a clear background comes out as a PNG with the same clear background, with no white fill, no halo, and no artifacts at the edges. If the WebP has soft semi-transparent areas such as shadows or anti-aliased text, those precise opacity values survive the round trip. This is the key reason to convert to PNG rather than JPG when the image has a transparent background, since JPG has no transparency channel and would fill those areas with a solid colour. ## WebP or PNG, which belongs where Keep WebP when the image lives on a web page. WebP is roughly 25 to 35 percent smaller than PNG at the same visual quality, so pages load faster and bandwidth costs less. Switch to PNG when the file needs to open in something offline. Photoshop versions before late 2022 require a plugin to open WebP. Word, PowerPoint, most legacy Windows tools, and the majority of email clients still refuse WebP. PNG is the format that every app understands. If your final destination is a website, keep WebP. If it is a design file, a print job, an office document, or an email attachment, PNG is the safer choice. ## Your output PNG will be larger than the WebP PNG uses exact compression. WebP uses a more modern and efficient algorithm that achieves smaller files for the same image. Going from WebP to PNG always grows your file, roughly 26 percent larger for graphics and two to ten times larger for photographs. A large photographic WebP can easily become several times heavier as a PNG. This is the cost of switching to a format that opens everywhere, since the PNG is not lower quality, it just takes more space because it stores every detail without the modern compression WebP uses. If file size matters more than compatibility, keep your WebP instead of converting. ## Where your WebP is processed It comes down to how many files you convert at once. Convert a single WebP and the re-encode runs inside the browser tab on the platform image engine, with nothing uploaded and nothing logged about the file. Open the Network panel in the developer tools and watch while you convert one image, and you see zero outbound requests carrying it. Convert several WebP files together and they are sent to our server, which produces one combined download for you. That result is removed from our server within about 2 hours, and you can delete it yourself as soon as it is saved. Most other WebP to PNG converters upload every file you give them, single or not, and keep it on their machines. Here a single conversion stays in the tab, and a batch reaches our server only long enough to assemble your download. ## What happens with animated WebP files Animated WebP files contain multiple image frames. This tool processes the first frame only and outputs a single still PNG. The PNG format supports animation through the APNG extension, but this tool does not produce APNG files. If you need to preserve every frame of your animated WebP, a dedicated GIF or APNG conversion tool is the right choice. For extracting a single representative frame, such as a thumbnail or a preview image, the first frame is typically the most useful output and the right default behaviour. ## Questions **Why can't Photoshop or Word open my WebP file?** WebP is a modern web format, and a lot of desktop software still ignores it. Photoshop only added native WebP support in late 2022, so older versions need a plugin. Word, PowerPoint, Preview on older macOS versions, and most legacy Windows tools simply refuse the format. Converting to PNG is the universal solution. Every application that handles images at all will open a PNG, making it the right choice when compatibility with desktop software matters. **Does this converter preserve transparency?** Yes, it is. Converting to PNG is the only right move if your image has a clear or semi-clear background and you need it to stay that way. JPG fills all clear areas with a flat colour, so the transparency simply disappears. PNG keeps it intact because the format carries the same transparency information that WebP stores. The practical test is straightforward: if a cut-out product photo, a logo, or any graphic with a clear background looked fine in the WebP, it will look identical in the output PNG. The opacity values are transferred exactly, with no edge halos and no pre-multiplied artifacts at soft edges like shadows. **Why is the PNG file larger than the WebP?** PNG is exact and WebP is more compressed. PNG stores every detail exactly, using a compression method from the 1990s. WebP uses modern prediction techniques that fit the same image into less space. Going from WebP to PNG always grows your file, about 26 percent for graphics and often two to ten times for photographs. A large WebP photo can become several times heavier as a PNG. This is expected, the price of a format that opens everywhere. **Does converting WebP to PNG lose quality?** The conversion itself introduces no new quality loss. PNG stores detail exactly, so the appearance decoded from the WebP is written faithfully. However, if your original WebP was saved in compressed mode, which most web WebP files are, those existing compression artifacts are already in the image and cannot be removed by converting to PNG. The PNG looks identical to how the WebP appeared on screen, not better. No quality is added, further quality loss is simply stopped. **Is it safe to convert WebP to PNG here?** It depends on how many files you convert. A single WebP is converted right on your machine, so that file is never uploaded. You can confirm it by opening the Network panel in the browser developer tools and watching during the conversion, where no outbound image request appears. When you convert several at once, they go to our server to be processed together, and that result is removed within about 2 hours, with the option to delete it yourself once you have downloaded it. If your image is sensitive, converting it one at a time keeps it on your machine. **What happens to animated WebP files?** Only the first frame is converted. This tool outputs a single still PNG, not an animated APNG, even when the source WebP contains multiple frames. If you need to preserve every frame, a dedicated GIF or APNG conversion tool is the appropriate choice. For most use cases, where you just need a static thumbnail or a preview from an animated WebP, the first frame is the right output. ## The details ### Why WebP is not universally supported in desktop apps WebP was released by Google in 2010, but adoption by desktop software lagged far behind browser support. Browsers adopted WebP early because they control their own rendering engines and Google actively optimized Chrome's WebP support as a competitive advantage. Desktop applications take longer to adopt a new image format because each application maintains its own image decoding stack, and supporting a new format requires testing against a long tail of edge cases. Adobe Photoshop, used by the majority of professional photographers and designers, did not ship native WebP support until version 23.2 in late 2022. Microsoft Office applications still handle WebP inconsistently across platforms as of 2026. Print RIP software and most legacy archiving tools never adopted WebP at all. The practical consequence is that a file that displays perfectly in a browser may be refused entirely by the software the user needs, and converting to PNG is the reliable workaround. ### How WebP achieves its compression advantage over PNG PNG uses a general-purpose exact compression method from 1996. It applies a set of reversible line filters before compression and encodes the result with a standard stream. This is effective but designed before video codec research produced better techniques. WebP lossy mode uses a block-based transform derived from VP8 video compression, applying intra-frame prediction and a frequency transform to large block units. WebP exact mode uses spatial prediction, colour transformation, and a coding stage that is structurally more efficient than PNG's approach for typical image content. Google's published benchmarks show exact WebP about 26 percent smaller than PNG on a standard set of test images, and compressed WebP with transparency about three times smaller than PNG at comparable visual quality. The PNG-to-WebP direction exploits these gains, the WebP-to-PNG direction reverses them, hence the larger output. ### Measured file size growth examples Measured on Chrome 148, Linux desktop, using the platform PNG encode path applied to decoded WebP inputs. A small vector-style graphic at a compact size, saved as WebP at quality 80, decodes and re-encodes as PNG in roughly 15 to 25 ms with a typical file size growth of 20 to 30 percent. A photographic WebP at a typical screen resolution decodes and re-encodes to PNG in under 100 ms with a typical growth of three to five times. A large photographic WebP at a high display resolution encodes to PNG in about 1.2 seconds with growth of five to ten times depending on scene complexity. These numbers reflect the difference in compression efficiency between the two formats and scale with overall image dimensions. ### Alpha transparency in the round trip The transparency channel in WebP and PNG uses the same value range, where zero means fully clear and the maximum value means fully solid. When the browser decodes a WebP with transparency, it produces an image buffer where the transparency component reflects the original data. When that buffer is re-encoded as PNG, the PNG encoder writes the transparency values into the PNG's transparency channel directly. No compositing step occurs, no background colour is applied, and no pre-multiplication side effects change the values. The result is a faithful transfer of the transparency channel, where every area's opacity in the PNG matches what the WebP stored. For images with fine anti-aliasing at edges, each intermediate opacity value survives the round trip intact. This fidelity is what makes PNG the right choice over JPG when the destination application needs to display the image on multiple backgrounds. ### EXIF and metadata behavior The re-encode pipeline strips EXIF, IPTC, and XMP metadata from the PNG output. WebP files can carry EXIF data embedded in their metadata chunk, and that data is lost when the browser decodes and re-encodes the image. ICC color profiles follow a different path, where Chrome and Safari preserve the sRGB ICC profile tag in the PNG output after decoding WebP, and Firefox strips all metadata including the ICC profile. The practical consequence is sRGB-safe output across all browsers, but any wide-gamut profile embedded in the source WebP does not survive in Firefox. For professional photographic workflows that rely on ICC-tagged round trips, use a metadata-aware conversion tool. For standard web image handling, metadata removal is typically acceptable and has the small benefit of slightly reducing output file size. ### Privacy verification in practice The claim that a single file is processed without uploading can be checked without any special tools. Open your browser, navigate to the webp-to-png page, then open the developer tools. Switch to the Network tab, clear any existing requests, and run a conversion by dropping a WebP file. Filter the request list by Fetch, XHR, or All. The list shows zero outbound requests containing image data during the encode. The only network requests present are the initial page load assets and standard analytics pings, which record page views and performance data only, with no image content. Every major remote WebP to PNG converter generates at minimum an upload POST and a download GET per conversion, both logged on the server. The local-first architecture means those log entries do not exist, which is the meaningful difference for users converting files that carry sensitive content. --- # About RoundCut Source: https://roundcut.app/about/ ## What RoundCut is RoundCut is a suite of free image tools. It covers the small, specific jobs people keep needing: cropping (round avatars, rectangles, social presets), converting between formats, compressing, removing backgrounds, resizing, AI upscaling, and turning images into PDF. Some of those run entirely in your browser, others route a single request through named infrastructure to do work the browser doesn't yet do as well, always with an invisible client-side fallback. No signup, no watermark, no resolution cap. The site exists because the alternatives, in our experience, get one of three things wrong: they upload everything to a server (whether or not the work needs one), they paywall basic operations, or they bury simple tools under intrusive ads. RoundCut tries to do none of those, and we describe the actual data flow honestly rather than claim "100% client-side" when the cloud path is what gives you remove.bg-grade results. ## Why we built it The first version of RoundCut launched in 2024 as a single tool, a circle crop for profile pictures, built to scratch a personal itch: the existing options either asked for an account or rendered the image at 256 pixels and offered a "Pro" upgrade for the original resolution. Neither felt acceptable for a one-line operation that every browser has been able to do natively for years. The site grew organically from there. Each new tool answered a recurring question we kept hitting personally, "how do I shrink this PNG without losing transparency?", "how do I strip the GPS coordinates from this photo before sharing?", "how do I make a Discord avatar without installing software?". The 2026 rebuild is the same project, rewritten on top of modern web standards (Astro, WebAssembly, Canvas API for the browser side. Cloudflare Workers + a small VPS + a cloud image API for the handful of tools where server-side encoders or AI models still beat the browser). ## Who makes RoundCut RoundCut is an independent project, not a company. It has been built and maintained by Guilherme since 2024, with AI tools helping in development and in writing the blog. The "we" on this site is one person, and that person reads every message sent to support@roundcut.app. ## Who it's for RoundCut is built for anyone who needs to do a small, specific thing to an image and doesn't want to install software, sign up for an account, or hand their photo to a generic uploader that retains it. The tools fall into a few categories of work: - **Cropping.** Circle-crop a photo for Discord, LinkedIn, Slack, Instagram, or any platform that masks avatars to a circle, plus rectangular and social-preset crops. Runs in your browser. Export as transparent PNG, WebP, AVIF, or JPEG. - **Format conversion.** Move between PNG, JPEG, WebP, and AVIF in either direction. Runs in your browser via Canvas and WebAssembly, except AVIF encoding, which routes one request to a Worker (with a main-thread browser fallback) because the in-browser AVIF encoder is slow on some devices. - **Compression.** Drop in one or more images, press Compress, and each file comes back smaller in its own format (JPEG, PNG, WebP or AVIF). The encode runs on our compression service, and falls back to an in-browser encoder if the service is unreachable. - **Background removal.** Pull a subject out of its background with AI. The default path sends your image once to a Cloudflare Worker that runs the BiRefNet model on Cloudflare's edge GPUs (same architecture as remove.bg), with the staging copy auto-deleted within one hour. When the cloud is unreachable, the tool transparently falls back to an in-browser model (ISNet via ONNX Runtime + WebAssembly), no upload at all on the fallback path. - **Resizing.** Change pixel dimensions with high-quality resampling, entirely in your browser. No upload. - **AI upscaling.** Enlarge an image with super-resolution. The default path uses a cloud model, with an in-browser resampling fallback when the cloud is unreachable. - **Image to PDF.** Wrap one or more images into a PDF. Runs in your browser for a single image, and routes to our service for multi-image documents, with a client fallback. Common to all of them: no upload that we keep, no account, no watermark. The server-touching tools document exactly what happens to your image, where, and for how long. ## Our principles ### Honest about the architecture "Your photos stay on your device" is a marketing line we deliberately don't use, because it's not true on every path. The truth is: the cropping, resizing, and format conversion (non-AVIF) work never leaves your browser, and you can verify that in DevTools' Network tab. The server-touching paths (background removal, compression, AI upscaling, AVIF encoding, and multi-image PDF) send your image exactly once, to named infrastructure we operate (a Hostinger VPS in Germany, Cloudflare Workers, and a cloud image API), do their work, return the result, and either auto-delete or rely on a content-addressed cache that's not tied to you. We tell you which path you're on, and we don't pretend the cloud path doesn't exist. If you need a hard guarantee that bytes never leave your device for a particular job, the browser-only tools cover that. The cloud-backed tools each ship an in-browser fallback that runs without any upload when you turn off the network. ### Free should mean free Every tool that exists today will stay free. We don't watermark exports, we don't cap resolution, we don't require an account, and we don't show interstitial ads. We may add a paid tier in the future for extra capability (batch processing, larger upload sizes, higher cloud quotas), but it would not gate anything that's already shipped. ### Truthful copy What we say a tool does and what the code actually does are the same thing. If the page says "Export as PNG, WebP, or JPEG", you'll find all three in the format picker. If the page says "up to 4096 px", that's the real cap in code. If a tool's default path is cloud-routed, the page says so, not "all-in-browser" with an asterisk. We have an internal rule that copy and code ship in the same commit, exactly so this never drifts. ### Accessibility from the start Every page passes WCAG 2.2 AA contrast in both light and dark themes. Every interactive element is reachable by keyboard, has a visible focus ring, and has a minimum 24 × 24 CSS-pixel touch target. Screen-reader landmarks (header, main, footer, nav) are in place. The site is built so that the experience for someone using a screen reader, a keyboard, or a small phone is not noticeably worse than for someone with a high-end desktop. ### No tracking by default We measure usage with our own cookieless beacon: anonymous event counts written to Cloudflare's Workers Analytics Engine, no cross-site identifier, and it honours your browser's Do Not Track setting. Cloudflare Web Analytics adds aggregate page-view counts, also without cookies. There are no third-party analytics scripts. Advertising cookies from Google AdSense load only on tool pages and, in regions that require it, only after you consent. The full data flow, including the cloud paths for the server-touching tools, is described in the [privacy policy](/privacy/). ## How we stay free RoundCut's static site runs on Cloudflare Workers (two Workers: one for the tools, one for the blog), and the browser-only tools cost us essentially nothing per use because the work happens on your machine. The server-touching tools have a small but real cost: - **Cloud processing** (background removal, AVIF encoding, AI upscaling) runs on Cloudflare Workers, an R2 bucket holding a few MB at any time, a KV namespace for rate-limiting, and a cloud image API for super-resolution, with per-IP and per-day caps that keep usage sustainable. - **The compression and PDF service** runs on a small Hostinger VPS shared with a sister project, returning sharp-grade encodes to thousands of requests per day. The cache hit ratio amortizes most of the work. Development is funded by advertising: - **Google AdSense on tool pages**: a band on the idle screen, a banner beside the editor on desktop, up to three units around the download screen, and two units inside blog posts. Institutional and legal pages stay ad-free. This is the standard pattern free tools like iLoveIMG, TinyPNG and Pixlr use. - **One sponsored card** on some download screens, labelled as such. If you buy through it we earn a commission and you pay the same price. Details in the [privacy policy](/privacy/). A paid tier, if we add one, would come only after sustained traffic, and free use stays free. ## How to reach us For questions, feedback, or bug reports, email [support@roundcut.app](mailto:support@roundcut.app). We read every message and try to respond within five business days. For privacy-related requests, see the [privacy policy](/privacy/). For a technical look under the hood, see [how it works](/how-it-works/). --- # Contact Source: https://roundcut.app/contact/ ## The fastest way to reach us Email [support@roundcut.app](mailto:support@roundcut.app). The same address handles every kind of request, there is no separate help desk, ticket portal, or contact form. A real person reads every message, usually within five business days. ## What to send where ### Support, a tool isn't working Tell us which tool you opened, what file you tried (format and approximate size), what you expected, and what happened instead. Browser name and version helps a lot (open `chrome://version` or the equivalent in your browser). Please don't attach the image itself unless we ask. RoundCut runs in your browser, so we usually can't reproduce a problem from your file. A screenshot of the error or the unexpected output is more useful than the source image. ### Privacy and data requests Right of access, deletion, correction, or any other request under GDPR, UK GDPR, LGPD, or CCPA: email the same address and put "Privacy request" in the subject line. Because RoundCut doesn't collect personal data tied to you, most requests are answered with written confirmation that we hold no record of you. We respond within the statutory deadline (30 days under GDPR / LGPD). ### Press and partnerships Interview requests, partnership proposals, integrations, or commercial inquiries: same email, "Press" or "Partnership" in the subject line. We're a small operation and prioritize partnerships that benefit users, please include a brief description of what you have in mind rather than a generic outreach template. ### Security disclosures Found a vulnerability? Email [support@roundcut.app](mailto:support@roundcut.app) with "Security" in the subject line, and please don't disclose publicly until we've had a chance to respond. We don't run a paid bug bounty, but we credit researchers who follow responsible disclosure on a public thank-you page when the fix ships. ## Who you're emailing RoundCut is a personal project run by an independent developer based in Brazil. The inbox isn't monitored 24/7, and there's no on-call rotation. If your message is urgent, say so in the subject line and we'll prioritize it. For questions specifically about how we handle your data, the [privacy policy](/privacy/) answers most of them in detail. For terms and acceptable use, see [terms of service](/terms/). ## Languages English and Portuguese are both fine. Other languages, we'll do our best with machine translation, but expect a slower reply. --- # How it works Source: https://roundcut.app/how-it-works/ ## The short answer RoundCut uses a **hybrid architecture**: most tools run entirely in your browser with zero upload, and a handful route a single network request through our own infrastructure when the browser cannot match the quality, always with an invisible client-side fallback. We tell you which path you're on, in every tool and on this page. The table below is **representative, not exhaustive** (new tools ship regularly). It shows one example of each kind of path: | Example tool | Where the work happens | |---|---| | **Circle Crop** (browser-only) | 100% in your browser, Canvas API. No upload, works offline. | | **Compress preview** (browser-only) | 100% in your browser, `canvas.toBlob`. No upload. The slider stays instant. | | **Compress final download** (server-touching) | One round-trip to our service at `api.roundcut.app` (Fastify + `sharp` + `libvips` on a VPS in Germany), with a browser fallback. | | **Remove Background** (server-touching) | One round-trip to a Cloudflare Worker that runs BiRefNet on Cloudflare's edge GPUs, with a WebAssembly fallback in your browser when the cloud is unreachable. | Cropping, resizing, format conversion (except AVIF encoding), and the compress preview all run fully in your browser. Background removal, the compress download, AI upscaling, AVIF encoding, and multi-image PDF each route a single request and fall back to the browser when the cloud is unreachable. You can verify the client-side claims in about 30 seconds: open DevTools → Network, clear the log, then use a browser-only tool like Circle Crop or the Compress preview slider, you will see zero requests carrying your image bytes leave the page. For the server-touching tools you will see exactly one upload per operation, to named endpoints. ## Why hybrid Most online image tools sit at one extreme: upload-everything-to-a-server (you wait for round-trips and the operator keeps your file), or all-in-browser (you pay quality and speed on the encode/AI steps). Neither extreme wins everywhere. We picked client-side where browsers are already excellent, the `<canvas>` element handles cropping, rotating, resizing, and the lossy preview encode in JPG/WebP, and server-side where the browser still measurably loses: - **Image compression**, on the final download. Server-side `sharp` + `libvips 8.17` produces files **10-15% smaller byte-for-byte** than browser encoders at the same visual quality, and gives access to AVIF speed/chroma tuning and JPEG XL output that the browser doesn't expose. The slider/preview still runs in your browser so iteration stays instant, only the "Download" tap goes through our service. - **AI background removal and AI upscaling**, on the default path. The models (BiRefNet for segmentation, super-resolution for upscaling) need a real GPU to finish in seconds. The in-browser fallbacks work but are slower and produce a visibly cruder result on hair, fur, and fine edges. - **AVIF encoding**, on conversion. The in-browser AVIF encoder is slow on some devices, so AVIF output routes one request to the compression Worker, with a main-thread browser encode as the fallback. The cost we accept for being server-side on those paths is one round-trip per operation. The cost we avoid by staying client-side everywhere else is the round-trip fee on the parts of the workflow that iterate fastest. ## The pipeline, step by step ### 1. You select a file Through the file picker, drag-and-drop, or paste, the browser hands JavaScript a `File` object. JavaScript reads the bytes using `FileReader` or `Blob.arrayBuffer()`. **At no point in this step is the file sent over the network**, regardless of which tool you're using. ### 2. The browser decodes the image Modern browsers natively decode JPG, PNG, WebP, GIF, and AVIF. We use `createImageBitmap()` to turn the raw bytes into a bitmap the GPU can work with, off the main thread. For HEIC on browsers that don't decode it natively, we fall back to a WebAssembly decoder that runs locally in your browser. ### 3. The tool does its thing, this is where the paths diverge - **Browser-only tools** (cropping, resizing, format conversion to PNG/JPEG/WebP, the compress preview). A Canvas 2D pixel transform or a WebAssembly codec re-encodes the bitmap locally. The result is ready without any upload. [Cropper.js](https://github.com/fengyuanchen/cropperjs) handles the interactive crop frame for the crop tools, and `canvas.toBlob` handles the preview re-encode. **Entirely in your browser.** - **Compress, download.** When you tap "Download", the image is sent once to `api.roundcut.app` (a Fastify service running on a VPS in Germany operated by Hostinger, Node 24 + `sharp 0.34` + `libvips 8.17`, the same C libraries Squoosh uses on its server path). It is re-encoded with the same parameters you set in the preview, and the bytes are streamed back to your browser. The service keeps a tenant-isolated, content-addressed cache (a hash of the input bytes + parameters) capped at 500 MB so that re-downloading the same image with the same settings replays the cached bytes, the cache is not indexed by you, IP, or filename. If the service is unreachable, the tool falls back to the in-browser preview blob. - **Remove Background, default cloud path.** The image is uploaded once to a Cloudflare Worker, staged in a private R2 bucket, processed by Cloudflare's `cf.image.segment` transformation which runs the **BiRefNet** model on Cloudflare's edge GPUs, and the cutout is streamed back. The staged R2 object is deleted within one hour by an R2 lifecycle rule, regardless of outcome. A typical photo finishes in 1-3 seconds. Daily per-IP and 5 MB upload caps keep the free tier sustainable. - **Other cloud-backed tools** (AI upscaling, AVIF encoding, multi-image PDF) follow the same shape: one request to named infrastructure (a cloud image API for super-resolution, the compression Worker for AVIF encode, our VPS service for multi-image PDF), result streamed back, and a transparent in-browser fallback if that request fails. - **WebAssembly fallback.** If the relevant cloud path is unreachable (your network drops, you're behind a strict firewall, the daily quota is full, or the file exceeds the cloud cap), the tool transparently switches to a local model or codec running in your browser, for example the **ISNet** background-removal model via [ONNX Runtime Web](https://onnxruntime.ai/docs/tutorials/web/) with WebAssembly. The first run downloads the model and is slower, subsequent runs are faster. **No upload on these fallback paths**, you can verify it in DevTools. ### 4. You download the result The output bitmap is encoded into a `Blob`, wrapped in an `object URL`, and offered to your browser's standard file-save dialog. The file appears on your disk. ## How to verify it yourself Pick whichever you prefer: ### Method 1. Watch the Network tab 1. Open RoundCut in a fresh tab and open DevTools → **Network**. 2. Use a **browser-only** tool, Circle Crop or the Compress preview slider. You will see requests for HTML/CSS/JS/fonts only, plus the relevant WebAssembly modules on first use. **No request will carry your image bytes.** 3. Now use a **server-touching** tool, Compress → Download or Remove Background. You will see exactly one `POST` carrying your image, and one response coming back with the result. Hover any request to see size and timing. The "Initiator" column tells you which script triggered each request, and the "Type" column tells you what was sent. We don't hide either. ### Method 2. Use the tools offline 1. Load any RoundCut tool page. Use Remove Background once on a small image so the in-browser ISNet model is cached. 2. Open DevTools → Network → tick **Offline** (or turn off Wi-Fi). 3. Reload the page, the static assets are cached, so it still loads. 4. Try the tools: - **Browser-only tools** (Circle Crop, resize, the Compress preview) keep working, they never needed the network. - **Compress Download** falls back to the in-browser preview blob (slightly less efficient encode, but functional). - **Remove Background** falls back to the ISNet WebAssembly model and works without any outbound request. If the tools worked offline (cloud-backed ones slightly degraded, browser-only ones identical), by definition no server saw your image. ## What we do see, and what we don't On the **client-side** paths, we see nothing about your image. There is no request to look at, no cache to store it in, no log line to grep. On the **server-side** paths: - **The compression and PDF service** sees the image bytes for the duration of the encode (typically a few hundred milliseconds), keeps a content-addressed cache entry for cache TTL, and that's it. The cache is not indexed by user, IP, filename, or any identifier we could use to find "your" images. We do not log image content. The encode service is shared across the same two tenants v1 served before cutover, with per-tenant CORS, rate limits, and HMAC-signed canonical URLs. - **Remove Background and the other AI paths** see the image for the duration of the staging upload and the model call (typically a few seconds), after which the staged copy is deleted by an R2 lifecycle rule. The BiRefNet model runs inside Cloudflare's own infrastructure, not on a remove.bg / fal.ai / Replicate-style external API. On every path, our analytics provider ([Cloudflare Web Analytics](https://www.cloudflare.com/web-analytics/)) records aggregate page-view data. URL, country, browser family, Core Web Vitals. No cookies, no persistent identifiers, nothing tied to a person. For tools that download a WebAssembly module on first use (the HEIC decoder, the ISNet ONNX model), our hosting provider sees that someone fetched the module, the same way it sees them fetch the CSS file. The module itself contains no information about your image. The full data inventory is in our [privacy policy](/privacy/). ## The technology stack For the curious: - **[Astro](https://astro.build/)**, the static site generator. Every page ships as plain HTML with progressively-enhanced JavaScript "islands" only where interactive tools live. - **Vanilla CSS with custom properties**, no Tailwind, no CSS-in-JS. The full design system is a single `tokens.css` file. - **`canvas.toBlob` / `<canvas>`**. JPEG, PNG, WebP, AVIF (browser-supported) encoding in the browser-only tools, and all of the crop and resize work. - **[Cropper.js](https://github.com/fengyuanchen/cropperjs)**, the crop-rectangle interaction layer. - **[ONNX Runtime Web](https://onnxruntime.ai/)**, runs the ISNet WebAssembly fallback for Remove Background. - **[Cloudflare Workers](https://workers.cloudflare.com/)**, host the static build as two Workers (tools and blog) and serve it from the edge. - **Cloudflare Workers + R2 + `cf.image.segment` (BiRefNet)**, the default Remove Background pipeline. - **Fastify + `sharp 0.34` + `libvips 8.17` on Node 24**, the compression and PDF service at `api.roundcut.app`, on a Hostinger VPS in Germany. - **Cloudflare Web Analytics**, aggregate, cookie-less page-view counts. ## Browser support Every tool works on the current and previous version of Chrome, Firefox, Safari, and Edge, desktop and mobile. The site uses progressive enhancement: where a browser supports a newer API (e.g. `showSaveFilePicker`, `OffscreenCanvas`), we use it, where it doesn't, we fall back to the older equivalent. There is no "your browser is not supported" wall. The only hard requirements are JavaScript (for any tool) and a network connection (only when using a cloud-backed tool on its default path, the browser-only tools run fully offline after the first page load). ## Questions Anything we didn't cover? Email [support@roundcut.app](mailto:support@roundcut.app). Technical questions welcome. --- # Privacy Source: https://roundcut.app/privacy/ ## The short version - **Most tools run 100% in your browser**. Circle Crop, Crop Image, Resize, single-image Convert and the Compress preview/edit step use the HTML5 Canvas API on your device. Nothing is uploaded. - **Some paths route your image through our infrastructure**: the final Compress download (when you click "Download"), Remove Background, Upscale, and the multi-file paths of Convert and Image-to-PDF. Your image is deleted after the job completes; it is never stored long-term, indexed, logged, or used to train anything. - **Whether a tool uploads can depend on how many files you pick.** Convert and Image-to-PDF run a single image entirely in your browser, and send a batch of two or more to our service. Each tool page states which path it uses. - **We do not set tracking cookies on our own surfaces.** We measure usage with our own cookieless beacon (an anonymous per-tab id in `sessionStorage`, no cross-site tracking). Third-party advertising cookies (AdSense) are loaded only on tool pages and only after you consent in regions that require it. - **You can verify the client-side claims yourself.** Open DevTools → Network, and watch Circle Crop, Crop Image or Resize run with zero upload requests. ## Who we are, controller identification For the purposes of GDPR (EU), LGPD (Brazil), CCPA/CPRA (California) and similar laws, the **data controller** is Guilherme Campos, reachable at [guilherme@roundcut.app](mailto:guilherme@roundcut.app) for data-protection requests. General support is at [support@roundcut.app](mailto:support@roundcut.app). Under Brazil's Resolução CD/ANPD nº 2/2022, this project qualifies as an "agente de tratamento de pequeno porte" and is not required to formally appoint a Data Protection Officer (Encarregado), the controller above serves as the direct communication channel with data subjects, as required by Art. 11 of that Resolução and §2 of Art. 41 of the LGPD. ## What information we process, and on what legal basis Each operation below is mapped to its legal basis under **LGPD Art. 7** (Brazil) and the equivalent **GDPR Art. 6** (EU). We declare these explicitly so you can audit which basis we rely on for which path. ### Your images. Browser-only tools When you use **Circle Crop**, **Crop Image**, **Resize**, or **Convert with a single image**, the work happens entirely in your browser via `<canvas>` and `canvas.toBlob`. The image bytes never leave your device. No version is sent to our infrastructure, cached, or logged, because no request carrying your image is made. **Legal basis: not applicable**, no processing by us occurs. ### Your images. Paths that route through our infrastructure Compress, Remove Background, Upscale, and the multi-file paths of Convert and Image-to-PDF route through our infrastructure. **Legal basis for all of them: LGPD Art. 7, V, execution of the user-requested contract (free service)**, plus our **legitimate interest (Art. 7, IX)** in operating the service sustainably. **Compress.** When you click "Compress", each image is uploaded once, through a signed URL, to a private Cloudflare R2 bucket and re-encoded in its own format by our compression service (`sharp` + `libvips` on a VPS in Germany operated by Hostinger; callers outside the Americas reach a regional instance of the same service on Cloudflare first). The result stays in R2 for up to two hours so you can download it, share it by link, or remove it at once with "Delete from server now"; after that it is deleted automatically. Neither the upload nor the result is indexed by user, IP, or filename, logged, used to train any model, or available to anyone outside the encoding service. If the service is unreachable, the tool encodes in your browser instead, and no upload happens on that path. **Remove Background.** Your image is uploaded once to a Cloudflare Worker (`roundcut-bg-remover`), staged in a private R2 bucket (`roundcut-bg-temp`), passed through Cloudflare's `cf.image.segment` transformation (which runs the BiRefNet model on Cloudflare's edge GPUs), and the resulting cutout is streamed back to your browser. The staged R2 object is deleted automatically within one hour by an R2 lifecycle rule, regardless of outcome. Uploads are capped at 5 MB and 30 requests per IP per day. If our Worker is unreachable, the tool falls back to an in-browser model (`onnxruntime-web`) that runs locally in your browser via WebAssembly, in that fallback path, no upload happens. **Upscale.** Your image is uploaded once to a Cloudflare Worker (`roundcut-upscale`) and enlarged by Cloudflare's `cf.image` `upscale=generate` transformation, which runs an AI super-resolution model on Cloudflare's edge. The staged object lives in the same private R2 bucket and is removed by the same lifecycle rule. If that path is unavailable, the image is instead uploaded to **ImageKit** (see Sub-processors) to be enlarged there. Per-IP daily limits apply. **Multi-file Convert and Image-to-PDF.** A single image in either tool is converted entirely in your browser and never uploaded. When you pick **two or more** files, they are sent to the same service as the Compress download (`api.roundcut.app`), processed there, and returned as a ZIP or PDF via a temporary download link that expires after about two hours. Each tool page states which of the two paths applies. In none of these cases is your image kept, indexed, sold, or shared with a third party beyond the named sub-processor that performs the encode or segmentation. We do not look at your images, and we do not have a workflow that would surface them to us. ### Site visit data We measure usage with our **own first-party beacon**, which writes anonymous event counts to Cloudflare's Workers Analytics Engine. It sets **no cookies**, does not fingerprint your device, and does not assign you a persistent identifier. It stores a random per-tab id in `sessionStorage` so repeated page loads in one visit are not counted as separate visitors; that id is discarded when you close the tab and cannot be used to recognise you later or across sites. The beacon honours your browser's **Do Not Track** setting: with DNT enabled, no event is sent at all. **Legal basis: LGPD Art. 7, IX, legitimate interest in measuring site usage in aggregate. GDPR Art. 6(1)(f) equivalent.** The data collected is limited to: the page URL, the referring URL, your country (derived from your IP, not stored individually), the browser family (e.g. "Chrome 130"), which tool was used and whether it succeeded, and Core Web Vitals performance numbers. None of this is linked to a person. ### Strictly-necessary cookies Cloudflare, as our hosting, DNS, and edge-compute provider, may set a small number of strictly-necessary cookies to deliver the site and protect it from automated abuse (for example, the `__cf_bm` bot-management cookie). These cookies are required for the site to function and are not used for advertising or tracking. **Legal basis: LGPD Art. 7, IX, legitimate interest in site security. Per the ANPD's 2022 Cookie Guide, strictly- necessary cookies do not require prior consent.** See the [Cloudflare cookie policy](https://www.cloudflare.com/cookie-policy/) for the full list. ### Local browser storage To make the site usable across visits, RoundCut stores small amounts of data in your browser's `localStorage`: your selected theme (light or dark), your preferred language, and a flag indicating whether you've dismissed the language-switch suggestion. We also keep a random per-tab id in `sessionStorage` for the usage beacon described above, which is discarded when you close the tab. The `localStorage` data never leaves your browser and is never visible to us. **Legal basis: LGPD Art. 7, V, execution of the user-requested service (your preferences).** ### Rate-limiting To prevent abuse of the Remove Background Worker, we keep a per-IP daily counter in a Cloudflare KV namespace bound to the Worker. The counter resets every 24 hours and stores only the count, not the images or any other content. **Legal basis: LGPD Art. 7, IX, legitimate interest in service availability.** ## Advertising RoundCut displays AdSense units served by Google on tool pages (a band on the idle screen, a banner beside the editor on desktop, and up to three units around the download screen) and two units inside blog posts. Institutional and legal pages remain ad-free. **Legal basis: LGPD Art. 7, I, consent of the data subject.** Per the ANPD's 2022 Cookie Guide, advertising cookies are **not** strictly necessary and require **explicit prior consent** with a banner that offers "accept" and "reject" with equal prominence. We use **Google Funding Choices** as our Consent Management Platform (CMP). It is AdSense's native CMP, implements the IAB TCF v2.2 consent signal, and is the [Google-recommended path for LGPD, GDPR and CPRA compliance](https://support.google.com/adsense/answer/13554116) for publishers serving ads. The CMP banner is shown automatically to visitors from regions that legally require prior consent (EU, UK, Brazil, California), your choice, accept all, reject all, or customize, is stored and respected on subsequent visits. You can re-open the consent settings at any time via the "Privacy" link that the CMP injects into the page footer when active. When you consent, third-party vendors including Google may use cookies and web beacons to serve personalized ads based on your visits to RoundCut and other websites. If you decline, AdSense still serves ads but only **non-personalized** ones, based solely on the page content and not on your browsing history. You may also opt out of personalized advertising globally by visiting [Google's Ads Settings](https://www.google.com/settings/ads), or [aboutads.info](https://www.aboutads.info/) for industry-wide opt-outs. ### Affiliate links Some download screens show one card labelled **"Sponsored"** that recommends a third-party service related to what you just did (for example, an AI headshot service on the circle-crop tool). Its link carries our affiliate code: if you buy there, RoundCut earns a commission and you pay the same price. The card is a plain link. Nothing from the partner loads on our pages, no cookie is set here, and we send the partner no personal data; any tracking starts on their site, under their own privacy policy. We count clicks on the card with the same cookieless beacon described above. ## Premium tier (future) We plan to offer an optional paid "RoundCut Pro" tier in the future, processed through [Lemon Squeezy](https://www.lemonsqueezy.com/) as merchant of record. If you choose to subscribe at that point, Lemon Squeezy will process your billing details (name, email, billing address, payment method) under their own privacy policy as a sub-processor. We will only receive a confirmation that your license is valid, plus your email address for licensing communication. **Legal basis: LGPD Art. 7, V, execution of the contract you enter into when subscribing.** Until the Pro tier launches, no payment information of any kind is processed by us or on our behalf. ## International data transfer Two of our sub-processors operate outside Brazil: - **Hostinger International Ltd.** runs the Compress download VPS in Germany. - **Cloudflare, Inc.** operates a global edge network, your Remove Background request may be processed at any Cloudflare data center, typically the one geographically closest to you. **Legal basis for the transfer: LGPD Art. 33, II, standard contractual clauses that Hostinger and Cloudflare publish for international transfers, combined with Art. 33, V, transfer necessary for the execution of the service the data subject requested.** GDPR equivalent: Art. 46(2)(c) SCCs. The ANPD has not yet (as of this update) published a list of adequate countries under Art. 34 of the LGPD, we will update this section if Germany or other operating jurisdictions are added to that list. ## Sub-processors - **Cloudflare, Inc.**. DNS, edge hosting (Cloudflare Workers static assets), DDoS protection, our analytics datastore (Workers Analytics Engine), Workers (`roundcut-bg-remover`, `roundcut-upscale`, `roundcut-compress-proxy`, `roundcut-metrics`), R2 storage (`roundcut-bg-temp`, auto-purged hourly; plus short-lived job results for batch downloads and shared links), KV rate-limit counters, the `cf.image.segment` transformation that runs BiRefNet, and the `cf.image` `upscale=generate` transformation used by Upscale. [Privacy policy](https://www.cloudflare.com/privacypolicy/). - **Hostinger International Ltd.**. VPS hosting in Germany for the image service behind `api.roundcut.app` (Compress download, multi-file Convert, multi-file Image-to-PDF). [Privacy policy](https://www.hostinger.com/privacy-policy). - **ImageKit (Raw Engineering Inc.)**. Fallback engine for Upscale: when the Cloudflare upscale path is unavailable, your image is uploaded to ImageKit to be enlarged and the result is streamed back. [Privacy policy](https://imagekit.io/privacy-policy/). - **Google LLC (AdSense)**, ad serving on tool pages, subject to user consent in regions that require it. [Privacy policy](https://policies.google.com/privacy). - **Lemon Squeezy (Lemon Squeezy LLC)**, only if you purchase a Pro license in the future. [Privacy policy](https://www.lemonsqueezy.com/privacy). ## Your rights as a data subject Depending on where you live, you have rights under laws including the EU's GDPR, **Brazil's LGPD (Art. 18)**, the California CCPA/CPRA, and similar regulations. ### Under Brazil's LGPD (Art. 18), you have the right to obtain from us: 1. **Confirmation** that we are processing your personal data (inciso I). 2. **Access** to the data we hold about you (inciso II). 3. **Correction** of incomplete, inaccurate, or outdated data (inciso III). 4. **Anonymization, blocking, or deletion** of unnecessary or excessive data, or data processed in violation of the LGPD (inciso IV). 5. **Portability** of your data to another service provider, subject to ANPD regulation and commercial / industrial secrets (inciso V). 6. **Deletion** of data processed under your consent, except for the cases the LGPD itself allows retention (inciso VI). 7. **Information about the public and private entities** with whom we have shared your data (inciso VII). 8. **Information about the possibility of not providing consent** and the consequences of that refusal (inciso VIII). 9. **Revocation of consent** at any time (inciso IX). To exercise any of these rights, email [guilherme@roundcut.app](mailto:guilherme@roundcut.app). Under the small-scale processing regime of Resolução CD/ANPD nº 2/2022 (Art. 14, I), our maximum response time is **30 days**, we aim to respond within **5 business days**. Because our service design holds essentially no personal data tied to you (no accounts, no persistent identifiers, no logged image content), most of these rights are satisfied by the design itself. If you would like written confirmation that no data is held about you, email the address above. ### Right to complain to the supervisory authority In Brazil, you may also file a complaint with the **Autoridade Nacional de Proteção de Dados (ANPD)** at [gov.br/anpd](https://www.gov.br/anpd/pt-br). In the EU, contact your national data-protection authority. ## Children and adolescents The LGPD defines a **child** as a person up to 12 years of age (complete) and an **adolescent** as 12 through 17 (Art. 14). RoundCut is a general-audience tool and is not directed at children. Processing data of a child requires **specific and prominent consent by at least one parent or legal guardian** (Art. 14 §1), and given that our service is not directed at children and does not request identifying information from any user, we do not knowingly collect children's personal data through normal use. If you become aware that a child has submitted personal information through the site, contact [guilherme@roundcut.app](mailto:guilherme@roundcut.app) and we will delete it. ## Changes to this policy If we make material changes, for example, when the Pro tier launches, or when we add a new tool that changes the data flow, we will update the "Last updated" date below and, where appropriate, post a notice on the home page. ## Contact {#contact} Questions about this policy or about how RoundCut handles data? Email [guilherme@roundcut.app](mailto:guilherme@roundcut.app). We aim to respond within 5 business days, the maximum legal response time under the LGPD small-scale processing regime is 30 days. **Controller:** Guilherme Campos. Brazil, [guilherme@roundcut.app](mailto:guilherme@roundcut.app) **Last updated:** 2026-05-15 --- # Terms Source: https://roundcut.app/terms/ ## 1. The basics These terms govern your use of RoundCut (the "service"), reachable at [roundcut.app](/). By using the service you agree to these terms. If you don't agree, please don't use the service. RoundCut is a personal project. The service is provided as-is, free of charge, with no warranty of any kind. We aim to keep the tools working and useful, but we can't promise uptime, error-free behavior, or that any specific result will work for your use case. ## 2. What you can do You may use every tool on RoundCut for any lawful purpose, personal or commercial , free of charge. We don't claim any rights over the images you process. Whatever license you have on the source image, you keep on the result. Two tools (Circle Crop and the Compress preview/edit step) run entirely in your browser and work offline after the initial page load. Two tools (the Compress final download and Remove Background) make a single network request to our infrastructure per operation, see §4 for the exact data flow, and therefore require a connection to complete a download. ## 3. What you can't do Some uses are off the table: - Don't use the service to process images you don't have the right to process. If you don't own the image, make sure you have permission. - Don't use the service to create content that's illegal where you live or where the affected people live, for example, content that's defamatory, infringes copyright, sexualizes minors, or is otherwise unlawful. - Don't try to break the service. No automated scraping at a rate that affects other users, no attempting to bypass security or rate limits, no probing for vulnerabilities outside a responsible-disclosure context. - Don't exceed the rate limits on tools that route through our infrastructure. Remove Background is capped at 5 MB per upload and 30 requests per IP per day, the Compress download service rate-limits per-tenant. Both limits exist to keep the free tools sustainable for everyone. - Don't redistribute RoundCut as your own product. The site, its design, its branding, and its source code are not in the public domain. ## 4. Your content and the data flow What happens to a file you process depends on which tool you use. We describe the full detail in the [privacy policy](/privacy/), the short legal version is: - **Circle Crop and the Compress preview/edit step** run entirely in your browser via `<canvas>`. We never receive these images. There is nothing for us to license, sell, or publish, because we never had it in the first place. - **The Compress final download** sends your image once to `api.roundcut.app` (a Fastify service on a VPS in Germany operated by Hostinger), re-encodes it with `sharp` + `libvips`, and streams the result back. We do not retain your image beyond a short tenant-isolated cache TTL, do not log it, do not use it to train any model, and do not make it available to anyone outside the encoding service. - **Remove Background** uploads your image once to a Cloudflare Worker, stages it in a private R2 bucket that is automatically purged within one hour, processes it via Cloudflare's `cf.image.segment` (which runs the BiRefNet model), and streams the cutout back. If our Worker is unreachable, the tool falls back to an in-browser WebAssembly model with no upload at all. In all four cases, we do not claim any ownership of your images, do not use them to train models, and do not share them with any third party beyond the named sub-processor that performs the encode or segmentation. The output of any tool is yours. RoundCut does not watermark exports and does not embed identifying metadata in the result. ## 5. Service availability and changes We may add tools, remove tools, change how an existing tool works, change the URL structure of the site, or take the service offline temporarily for maintenance. The infrastructure paths described in §4 (`api.roundcut.app`, the Remove Background Worker, the R2 bucket) may also be replaced with equivalent providers in the future, when we do, we update §4 and the privacy policy in the same commit. When we make a meaningful change to a tool's capability or to these terms, we'll update the relevant page. Significant policy changes will be announced on the home page. We don't guarantee that any specific tool, format, feature, or hosted endpoint will remain available forever, although we try not to remove things people rely on. ## 6. Premium tier (when launched) We plan to introduce an optional paid "RoundCut Pro" tier in the future, processed through Lemon Squeezy as merchant of record. Until that tier launches, the entire service is free with no exceptions. When Pro launches, the additional terms covering billing, refunds, and license validity will be added here in the same commit that activates the tier. Free tools that exist today will remain free indefinitely. Pro adds capability, it doesn't gate what's already shipped. ## 7. Advertising RoundCut displays AdSense units served by Google on tool pages (a band on the idle screen, a banner beside the editor on desktop, and up to three units around the download screen) and two units inside blog posts. Institutional and legal pages remain ad-free, and Pro subscribers (when the tier launches) will not see ads at all. We have no control over the specific ads served by AdSense, although Google operates AdSense under its own content policies. If you see an ad you believe violates Google's policies, you can report it through Google's tools. ## 8. Ownership and third-party names The site's design, copy, branding, and custom illustrations belong to the project. The site's source code is private. Third-party names, logos, and trademarks mentioned on the site (e.g. "Discord", "LinkedIn", "Instagram", "iPhone", "HEIC", "Cloudflare", "Hostinger", "sharp", "BiRefNet") belong to their respective owners and are referenced only descriptively, to explain what the tools do and which providers power them. ## 9. Disclaimer of warranties The service is provided "as is" and "as available", without warranty of any kind, express or implied. We don't guarantee that the tools will produce a specific result, that they will work in every browser, that they will be available at any specific time, or that they will be error-free. To the maximum extent permitted by applicable law, we disclaim all implied warranties, fitness for a particular purpose, merchantability, non-infringement, accuracy. ## 10. Limitation of liability To the maximum extent permitted by applicable law, RoundCut and its operator are not liable for any indirect, incidental, special, consequential, or punitive damages arising out of your use of the service, including loss of data, lost profits, or business interruption, even if we've been advised of the possibility of such damages. Our total cumulative liability arising from these terms is limited to USD 100, or the amount you've paid us in the past twelve months for the Pro tier (whichever is greater). For free use, our cumulative liability is limited to USD 100. ## 11. Indemnity You agree to indemnify and hold us harmless from claims, damages, or costs (including reasonable legal fees) arising out of your use of the service in violation of these terms, for example, processing content you don't have the right to process, or using the service for an illegal purpose. ## 12. Privacy How we handle data is described in detail in our [privacy policy](/privacy/). The summary: two tools run entirely on your device with zero upload, two tools (Compress download and Remove Background) route a single request through our infrastructure, do not retain your image, and do not use it for training. ## 13. Governing law These terms are governed by the laws of Brazil, without regard to conflict-of-laws principles. If a dispute arises that we can't resolve informally, the courts of São Paulo, Brazil will have exclusive jurisdiction, except where local consumer-protection law gives you the right to bring a claim in your own jurisdiction. ## 14. Contact Questions about these terms? Email [support@roundcut.app](mailto:support@roundcut.app). ---