How to edit an image online
Drop or paste your picture, pick a tool on the left or a command in the menus above it, and press Save. Every change can be undone with Ctrl+Z.
Open a picture
Drop it on the box, choose it, or paste a screenshot with Ctrl+V. A photo opens on Adjust, a screenshot or logo on Move.
Crop and straighten
Drag the corners, or pick a ratio such as 1:1, 4:5 or 16:9, or a size for a post. Auto levels a tilted horizon.
Make the changes
Fix the light and colour, add text and arrows, paint, hide names and faces, select something and copy it to a layer, or let the AI remove a background or an object.
Save or copy
PNG, JPG or WebP from the arrow on the Save button. Save options sets a largest file size for forms with a limit; Ctrl+C copies the result into a chat or a document.
Layers
The Layers panel under the tool's settings lists every layer, the top one first. Choose one to edit it; drag a row to move it up or down the stack; the eye hides it. Each layer has its own opacity and blend mode, as in desktop editors.
- Text and shapes are layers of their own and stay editable: move them, change the words, the colour or the size, until you turn them into pixels
- A picture you paste or drop on an open one becomes a new layer, made smaller to fit when it is larger
- Adjustments stay settings on their layer, so they can be changed or reset later without losing quality
- Crop, turn, resize and canvas size change every layer at once; saving flattens what you see into one file
What each tool does
| Move | Drag a layer, nudge it with the arrow keys, scale it by a corner, set its exact place, or line it up with the picture's edges or centre |
|---|---|
| Select | Rectangle, ellipse, lasso, painting, by colour, or an object found by AI with a click. Then copy or cut it to a new layer, delete, fill, hide, remove it with AI, or crop to it |
| Crop | Free or fixed ratio, sizes for posts (Instagram, Stories, YouTube thumbnails, link previews, headers), exact pixels, a rule-of-thirds grid; straighten up to 45° with Auto; turn and flip without losing quality |
| Adjust | Auto fix, exposure, contrast, highlights, shadows, warmth, tint, saturation, vibrance, sharpen and grain on the chosen layer, with a before and after divider; Black and white and Invert colours in the Adjust menu |
| Brush | Paint or erase with a size, hardness and opacity; Shift + click draws a straight line |
| Text | Text in four fonts (outlined, plain or as a label), arrows with one or two heads, lines, boxes and ellipses (outlined, tinted or solid), a marker that keeps text readable, and numbered steps for how-to screenshots |
| Hide | Solid in any colour, pixelate or blur, as a rectangle or an ellipse; Hide every face finds the faces for you |
| Remove | Paint over a person, a sign or a wire and AI fills the area from what is around it; Try another fill gives a different result |
| Colour picker | Pick any colour as HEX, RGB or HSL (one pixel or an average), check the WCAG contrast of two colours, and get the picture's palette as CSS |
| Image menu | Resize (smooth, or sharp pixels for pixel art), canvas size with an anchor and a fill, perspective (to A4, Letter, a screen or an exact size), trim edges, make square, make a colour transparent |
| Save | PNG, JPG, WebP (and AVIF where the browser writes it) with quality, a largest file size and a file name, or copy to the clipboard |
AI tools inside the editor
The site's AI picture tools work on the chosen layer, with their own settings, and their results are steps you can undo like any other. They run on your device; each downloads its model from this site the first time, then keeps it.
| Remove background | Cuts out the subject; the original can stay as a hidden layer under it. On its own page: Background remover |
|---|---|
| Remove an object | Fills painted areas, or a selection. On its own page: Remove an object |
| Select an object | Click a person, an animal or a product to select it, with clean edges. On its own page: Cut out an object |
| Upscale | 2×, 3× or 4× with real detail, every layer with it. On its own page: Image upscaler |
| Colourise | Natural colour for black and white photos. On its own page: Colourise old photos |
Hiding private details safely
Blurring or pixelating text looks safe but is not always: with small blocks, the original can sometimes be recovered by trying fonts until the blocks match. For passwords, card numbers, addresses and anything else in text, use Solid. It replaces the pixels with one colour, and the saved file has nothing underneath.
- Solid is the default, in black, white, grey or any colour
- Hiding works through every layer, so nothing hidden is left on a layer underneath
- Pixelate uses large blocks; blur is heavy and has noise added, so it cannot be sharpened back
- Hide every face finds faces with YuNet, on your device (small ones too), and covers each with an oval sized so the finder cannot see a face in it any more
- The saved copy has no location, camera or date in it, even when the original did
Where does the picture go?
It stays on your device. Every edit happens right in your browser. The AI tools download their models from this site the first time you use them (the face finder is 14 MB, the background remover about 105 MB); Models on this device lists and deletes them.
Specs and shortcuts
| Input | JPG, PNG, WebP, GIF (first frame), BMP, AVIF, up to 50 megapixels. Orientation from the camera is applied |
|---|---|
| Output | PNG, JPG or WebP. Transparency kept in PNG and WebP; JPG gets a white background |
| Undo | Every step, as far as memory allows (the oldest steps of very large pictures are let go first); History jumps to any step |
| Tools | V move, M select, C crop, A adjust, B brush, T text, H hide, R remove, I colour picker; [ and ] change the brush size; Ctrl+J with a selection copies it to a new layer |
| Keyboard | Ctrl+Z undo, Ctrl+Shift+Z or Ctrl+Y redo, Ctrl+S save, Ctrl+Shift+S save options, Ctrl+C copy, Ctrl+V paste as a layer, Ctrl+A select all, Ctrl+D deselect, Ctrl+J duplicate the layer, Ctrl+E merge down, arrows nudge, hold \ to see the original, hold Space to pan, + and − zoom, 0 fits, 1 shows actual pixels, Enter applies a crop, Esc cancels |
| Browsers | Current Chrome, Edge, Firefox and Safari, on computers and phones |
| Price | Free, no watermark, no account |
Questions
No. Every edit runs in your browser, on your device. It stays on your device, and the saved copy leaves out the location, camera and date that photos carry.
Not always: pixelated or lightly blurred text can sometimes be read back by trying fonts until the blocks match. For text, use Solid, which replaces the pixels with one colour. The editor uses large blocks and adds noise to blurs, but Solid is the safe choice.
Open the logo and choose Image, Make a colour transparent. Pick the background colour and press Make transparent. Smooth edges un-mixes the old background out of the soft edge pixels, so the logo has no white halo on a dark page. For a photo, AI, Remove background cuts out the subject instead. Save as PNG or WebP.
Yes. In File, Save options, choose a largest size such as 500 KB or 1 MB. The editor finds the highest JPG or WebP quality that fits, and makes the picture smaller only if quality alone is not enough.
Yes. Each layer has its own opacity and blend mode, and can be hidden, renamed, reordered, merged or flattened. Text and shapes are layers that stay editable, and a picture you paste or drop becomes a new layer. Saving flattens what you see into one PNG, JPG or WebP.
For most everyday jobs, yes: layers, selections, text, a brush, crop and colour, and AI that removes backgrounds and objects. It does not open or save PSD files, and has no masks, adjustment layers, paths or CMYK.