Free Online Image Tools
Edit photos in your browser. No upload, no quality grab.
Photos are personal. SwiftTooly's image tools use the browser's native Canvas API to compress, resize, crop, convert and inspect your images locally — so a screenshot of a passport or a private family photo never gets parked on someone else's hard drive. Drag a file in, get a new file out, and close the tab when you're done.
8 Image Tools
Why use SwiftTooly's Image Tools
Your photos stay yours
No upload pipeline, no analytics on the pixel data, no AI training set.
Format-preserving
JPG stays JPG, PNG stays PNG — no surprise re-encodes or stripped metadata.
Pixel-precise
Type exact dimensions, exact quality, exact crop coordinates. No guessing.
Batch-friendly speed
Operations are local, so opening a few dozen files in a row stays snappy.
Image Tools in your browser vs uploaded to a server
SwiftTooly (browser)
Images are decoded into a <canvas>, edited with HTMLCanvasElement APIs, and re-exported as a Blob — all inside the tab.
Typical upload-to-server tools
Your images are uploaded to someone else's storage, processed by their pipeline, retained in their cache for hours or days, and may be used to train future models.
Should you compress, resize or convert an image?
These three overlap enough that people frequently reach for the wrong one. Resizing changes pixel dimensions — the actual width and height of the image. Compressing keeps the dimensions and reduces file size by discarding image detail. Converting changes the container format, which may or may not change the size depending on which formats are involved. If a photograph is too large as a file, resizing is almost always the more effective first step, because file size scales with pixel count. A 4000-pixel-wide photo shown at 800 pixels on a web page carries twenty-five times more pixels than it needs, and resizing to 1600 removes most of the weight before any compression is applied. Compress afterwards to fine-tune. Reaching for compression alone on an oversized image means throwing away quality to solve a problem that resizing would have solved without any visible loss.
A sensible order of operations
When several transformations are needed, the order changes the result. Crop first, because cropping discards pixels and there is no point processing pixels you are about to remove. Resize second, to the final display dimensions you actually need. Convert third, to the format the destination wants. Compress last, because compression artefacts introduced early get amplified by every subsequent operation that re-encodes the image. The reason this matters is generation loss. Every lossy encode discards information permanently, so an image that is compressed, then resized, then compressed again has been through two lossy passes and carries the damage of both. Keeping a lossless master — the original file, or a PNG — and generating each delivery copy from that master in a single pass produces noticeably better results than editing a delivery copy repeatedly.
Formats, transparency and the mistakes that cost most
The single most common unpleasant surprise is converting a transparent PNG to JPG and finding a white box behind the logo. JPG has no alpha channel and cannot represent transparency at all, so the transparent area is composited onto a background. If you need a smaller file and transparency, WebP does both. The second most common is under-sizing for high-density screens. An image displayed at 400 pixels wide on a modern phone or laptop needs roughly 800 pixels of actual data to look sharp, because those screens pack two or more physical pixels into each CSS pixel. Exporting at display size is why images look soft on retina displays. The third is printing from a web-sized image. Print quality is measured in dots per inch and 300 dpi is the standard, so a 6-inch-wide print needs 1,800 pixels. An image that looks perfect on screen at 800 pixels will look visibly soft printed at any reasonable size. A fourth worth knowing about is metadata. Photographs from phones carry EXIF data including GPS coordinates, and redrawing an image through a canvas — which is what resizing, cropping and converting all do here — does not carry that metadata into the output. That is a useful side effect when sharing photos publicly, and something to be aware of if you were relying on the capture date or camera settings being preserved.
Related collections
Frequently asked questions
Do you keep a copy of the photos I edit?
- No. Image data lives only in your browser's memory while you work and is discarded when you close the tab. We do not see, log or store the pixels.
Will compression visibly reduce quality?
- It depends on the quality slider. Most photos look identical to the eye at 70–85% quality while shrinking to a fraction of the original file size.
Which image formats can I work with?
- Anything your browser can decode — JPG, PNG, WebP, GIF and (on modern browsers) HEIC. The Image Converter tool also handles cross-format conversions.
Does resizing preserve transparency?
- Yes. PNG and WebP transparency is preserved through resize, crop and convert operations. JPG, by definition, has no alpha channel.