Preset, dimensions, output format, and compression.
Quick Preset
Optimize for
Smart Auto
Website
WordPress
Email
Facebook / OpenGraph
Instagram Square
Instagram Portrait
Compress only
Custom
Smart Auto analyzes each file and picks practical web settings automatically.
✦ Per-image format, size and quality decisions
Saved custom presets
Choose a saved preset…
Save current
Delete saved
Crop, Rotate & Flip
Crop / aspect ratio
No crop Square · 1:1 Portrait · 4:5 Photo · 3:2 Wide · 16:9 Facebook / OG · 1200:630
Crop focus
Smart · keep the interesting part Center Top Bottom Left Right
Rotate 0° 90° clockwise 180° 270° clockwise
Flip None Horizontal Vertical Both
For precise framing, use the Crop button beside an image. You can drag and resize the crop area visually for each image.
Colour
Fix it for me
Leave it alone
Applied to every image in the batch. "Fix it for me" reads each picture's own
brightness spread and stretches it, so a flat or dark photo is corrected per image
rather than all by the same amount.
Enlarge & Enhance
Enlarge by
No enlargement
2×
3×
4×
Enhancement
None
Enhanced
AI Enhance
Image type
Photo
Graphic / Logo
Screenshot / UI
Text / Document
Reduce noise / compression grit
Sharpen after enlargement
AI Enhance uses an ESRGAN neural network in your browser. The self-hosted model loads only when AI Enhance is used, then the image is processed locally. AI is 2× internally; 3×/4× outputs use AI detail reconstruction first, then a high-quality final resize.
For a specific custom size, leave Enlarge by at “No enlargement” and use the Resize fields above.
Background
Background treatment
Keep original background
AI remove → transparent
AI remove → solid color
AI remove → blurred original
AI remove → replacement image
Removal quality
Fast · U²-Net P (~5 MB model)
Higher quality · U²-Net (~176 MB model)
Replacement color
U²-Net loads only when this feature is used. The model runs directly through ONNX Runtime in the visitor's browser.
Test Background AI
Faces
Hide every face in every picture
How to hide them
Blur
Pixelate
Black box
A detector finds most faces, not all of them. Check the results before you
publish anything where a miss matters. Faces turned away, very small, or in
deep shadow are the ones it loses.
It is baked into the pixels, like the other hiding tools, so it cannot be
undone from the file you download. The first picture also downloads a 1.2MB
detector; after that it is instant.
Convert
Output format
WebP
JPG
PNG
AVIF
Keep original format
This browser cannot encode AVIF with Canvas. Choose WebP/JPG/PNG instead.
JPG does not support transparency. Transparent areas will use the background color below.
JPG background
Compress
Optional target size per image (KB)
PNG export is lossless, so the quality slider and target-KB setting cannot reliably shrink it.
Compression and dimension goals can be used together — for example 1600px wide and under 250 KB. Metadata behavior is under Advanced.
Watermark
Watermark type None Text Logo / image
Multiple Outputs
Output recipe
Single output
Web bundle · WebP + JPG
Social pack
Favicon / app icons
Responsive website set
Custom multi-output set
Creates one output using the Basic settings.
Metadata & Orientation
Output metadata Strip all metadata Keep copyright / artist when possible Keep EXIF on JPG outputs
Normalize phone / camera orientation before processing
Full EXIF reinsertion is supported for JPG → JPG. Other formats can be inspected, but Canvas-based WebP/PNG/AVIF exports do not reliably preserve full EXIF.
Clipboard
Finished images get a Copy button. Clipboard image writing requires HTTPS and browser support; unsupported formats are converted to PNG for copying.
Keep it on your computer
This can be installed as an application on a Mac or on Windows, in its own window and
working with no connection. How that works, and how to do it
by hand .
Checking whether your browser can install it…
Install it
How ImageReady works
A free image compressor, converter, resizer and enhancer that runs entirely in your browser. Common questions:
Are my images uploaded anywhere? No. Every step runs inside your browser using the Canvas API and WebAssembly. The only network requests are for this page and its own code, encoders and AI models, which are all served from this site. Close the tab and nothing is left behind.
Which image formats can I use? Input: JPG, PNG, WebP, AVIF, HEIC/HEIF from iPhones and the first frame of a GIF. Output: WebP, JPG, PNG or AVIF. AVIF and MozJPEG use WebAssembly encoders bundled with the site, so they work in every current browser.
How do I get an image under a certain file size? Open Basic and enter a target size in the optional KB field. The tool re-encodes at decreasing quality until the file fits, and you can combine that with a maximum width, for example 1600 px wide and under 250 KB.
What does Smart Auto do? It looks at each image on its own, including dimensions, transparency and file size, and picks a sensible web format, maximum dimension and quality. Choose another preset or Custom to take control of those settings.
Can I process a whole folder at once? Yes. Drop many files or choose a folder, apply the same settings to all of them, and download the results as one ZIP. The Advanced tab can keep your folder structure inside the ZIP and skip exact duplicates.
What do the AI features download? AI Enhance and background removal also run locally. The first use downloads the model to your browser: about 2 MB for AI Enhance, about 8 MB for fast background removal and about 176 MB for the high-quality model. Nothing is sent to a server.
Does it remove metadata like location and camera details? By default all metadata is stripped from the output. In the Advanced tab you can keep copyright and artist fields, or keep full EXIF for JPG to JPG conversions. Phone photos are rotated correctly before processing.
Is it free, and which browsers work? It is free with no account and no file limits beyond your device's memory. Use a current version of Chrome, Edge, Firefox or Safari on desktop or mobile. AVIF encoding is CPU-heavy, so expect a few seconds per image.
Before / After
×
Original: Processed:
Zoom Fit 100% 200%
Edit
×
Crop
Mark up & hide
Angle
Colour
Fix this one automatically
Back to the batch setting
Following the batch setting in the panel on the right. Move a slider and this picture
gets its own.