Image

PNG is lossless — re-saving never degrades it — and supports transparency. It suits screenshots, diagrams, logos, and anything that needs a see-through background. The trade-off is that photographs saved as PNG tend to be large.

JPEG is lossy and excels at photographs with smooth gradients. It has no transparency and loses a little quality every time it is re-encoded, but it is the safe default for sharing or attaching photos.

WebP does both lossless and lossy, supports transparency, and is usually smaller than JPEG or PNG at comparable quality. It is the first choice for lightening images destined for the web, which is why the format converter defaults to it, with quality preset to 90 and adjustable on a slider. Be aware that some older software still cannot open WebP, so for a file you are handing to a specific person, JPEG or PNG is the safer bet.

Photos taken on an iPhone are often stored as HEIC. It is space-efficient, but older Windows setups and plenty of web services cannot open it. HEIC Converter turns those files into PNG or JPEG — pick JPEG for sharing a photo, PNG if it is an intermediate step in further editing.

To turn a set of stills into a short animation, use Image to GIF. Drag the frames into order and set the playback speed. GIF caps out at 256 colours, so photographs can show visible banding; illustrations and screen recordings fare much better.

When you want a lighter image, resizing and compressing attack the problem from opposite directions. Resizing removes pixels: take a 4000 × 3000 photo down to 1200 × 900 and there is simply far less data to store. If you are shipping an image larger than the space it will be displayed in, resizing is by far the biggest win, and you can keep the aspect ratio locked while you do it.

Compression keeps the pixel count and lowers the encoding quality instead. For JPEG and WebP a quality slider controls that, with a preview so you can see what you are giving up. PNG works differently: it is reduced by cutting the palette to 64, 128, or 256 colours. Illustrations and screenshots survive this almost untouched, while images full of gradients may show banding.

In practice the efficient order is: resize down to the size you actually need, then compress to finish. Both tools accept multiple files at once and hand the results back as a ZIP.

Photo files carry EXIF metadata: capture time, camera make and model, exposure settings, and on many devices the GPS coordinates of where you stood. Some social platforms strip it on upload, but send the file by email or chat and it almost certainly travels with the photo. Remove EXIF redraws the image on a canvas and re-encodes it, dropping that metadata along the way.

To see what is in there before you publish, run Privacy Checker. Alongside EXIF and GPS it detects QR codes and barcodes visible in the picture — using the browser's BarcodeDetector for a wide range of formats on Chrome, Edge, and Android, and falling back to QR-only detection on Safari and Firefox. It also flags the reduced-size thumbnail some cameras embed inside EXIF, because an unedited thumbnail can reveal exactly what you thought you had covered up.

When a face, an address, a name, or a company logo is in frame, Image Markup lets you lay down blackouts or blur. The work is done by rewriting the actual pixels on a canvas, so the exported image no longer contains the originals. Unlike a PDF blackout, there is nothing underneath to recover. Do note that blur leaves some of the original shape and colour behind — for anything you truly need hidden, use a solid blackout. Seal stamps made in the Seal Stamp Maker can be pasted in with Ctrl+V.

When something unwanted sits at the edge of the frame, or you only need the middle, Crop keeps just the region you want — free-form or with the aspect ratio locked, which makes square profile and thumbnail crops painless. Sideways photos can be fixed with Rotate & Flip: 90-degree rotations plus horizontal and vertical mirroring.

Background Removal estimates the subject's outline with an AI model, makes the background transparent, and exports a transparent PNG. It uses a compact model (u2netp, about 4.4 MB) that is downloaded to your browser and run on your device — good for ID photos, profile pictures, and product shots. Being a lightweight model, it can soften the boundary around fine hair or where subject and background share similar colours. It runs in Chrome, Edge, Safari, and Firefox alike, though slower devices take longer.

Color Palette Extractor pulls the dominant colours out of an image and hands them to you as hex values or CSS variables, which is handy when you want a site or a deck to match a photograph. Retro Effect combines sepia, film grain, blur, tone adjustment, and a white border to give a picture a film-photo feel.

Resizing, cropping, rotating, masking, and retro effects are all done by redrawing the image through the browser's built-in Canvas API. Format conversion and compression simply re-encode that canvas at the format and quality you chose. No plugin is involved.

Metadata is read with exifr, codes are detected with BarcodeDetector and jsQR, GIFs are built with gifenc, HEIC is converted with heic2any, and background removal runs on ONNX Runtime Web. Every model and WebAssembly file is served from our own origin — there is no third-party CDN and no external inference API.

Because the pixels never leave your device, the files you would hesitate to upload — internal screenshots, a photo of an ID, family pictures — can be processed as they are. Close the page and the in-memory data is gone, with no history left behind.

OFFiler runs as a PWA, so once it has loaded you can open the tools and process files with no connection at all — resizing a photo or stripping its EXIF on a train or a plane works fine.

Output images are saved to your device's Downloads folder. No intermediate result is ever left on a server.