About this tool
Apply one reproducible resize and encoding policy across a batch, inspect each file result, and retain an export manifest for publishing or QA.
Batch Image Optimizer applies one resize and encoding policy to up to 30 images at once and hands back the results as individual files or a ZIP with a JSON manifest. It reads PNG, JPEG, WebP, GIF, and AVIF sources up to 20 MB each and 300 MB per batch, draws each one onto a canvas at the size you request, and encodes it as WebP, JPEG, or PNG. Every output is decoded again so its MIME type, signature, and pixel dimensions can be checked before you download, and the file list shows before-and-after bytes for each image. The typical job is preparing product photos, blog illustrations, or screenshots for a website where consistent dimensions and a smaller payload matter more than pixel-perfect fidelity.
- Validates PNG, JPEG, WebP, GIF, and AVIF signatures, decoded dimensions, 20 MB file limits, and a 30-file or 300 MB batch ceiling.
- Supports contain, cover, and long-edge resizing with bounded target canvases, alpha or solid padding, and explicit upscaling control.
- Processes sequentially with cancellation between files and verifies MIME, signatures, and decoded dimensions for every output.
- Exports collision-safe files plus a reopened ZIP and JSON manifest recording settings, dimensions, formats, and before/after bytes.
How to use Batch Optimizer
Click Add image files or drop images onto the panel; each is checked for a real image signature and decoded dimensions before joining the list, and a per-file remove button drops strays. Pick a Resize mode: contain pads the whole image inside the target canvas, cover fills the canvas with a centred crop, and long-edge keeps the aspect ratio and scales the longest side to one number. Enter Width and Height, or Long edge, up to 8,192 px and 24 megapixels. Choose an Output format, drag the Quality slider between 30% and 100% for WebP or JPEG, and set Transparent padding or a Background color. Toggle Allow upscaling if small sources should grow. Press Optimize, then download files one by one or click Download ZIP + manifest.
When this tool is useful
- A shop owner converts 25 supplier JPEGs to 1600 px WebP files before uploading them to a product catalogue.
- A blogger resizes screenshots taken on a Retina display to a 1200 px long edge so posts stop shipping 4 MB images.
- A designer produces square 512x512 thumbnails from mixed portrait and landscape photos using cover mode.
- A developer needs PNG icons padded to an exact 256x256 canvas with transparency preserved for an app build.
- A newsletter editor batch-converts AVIF images from a photographer into JPEG because the email client cannot render AVIF.
Practical tips
- EXIF, GPS, and colour profiles are stripped by canvas re-encoding, and animated GIFs are flattened to one frame; keep originals if that matters.
- Leave Allow upscaling off unless you need exact dimensions; enlarging an 800 px photo to 1600 px only adds bytes and blur.
- WebP at 80-85% is a sensible default for photos. For flat-colour UI screenshots, PNG is often smaller and stays lossless.
- JPEG cannot hold transparency, so contain-mode padding uses the Background color; check it matches your page background.
- Files are processed one at a time, and Cancel after current file lets you stop a long batch without losing finished results.
Examples you can test
Load an example, compare the result with the expected output, then replace it with your own input.
Product photos to WebP
Example input
18 JPEGs between 3 and 6 MB; long-edge 1600 px; WebP; quality 82%; upscaling off
Expected output
18 WebP files a fraction of the original size, a ZIP named optimized-images.zip, and manifest.json listing settings and per-file dimensions
The manifest records both source and output bytes, which makes it easy to show a reviewer how much the page payload dropped.
Square thumbnails with a crop
Example input
Mixed portrait and landscape PNG photos; cover mode; 512 by 512; JPEG; background #ffffff
Expected output
512x512 JPEGs with the centre of each photo kept and the edges cropped away
Cover crops symmetrically from the centre, so subjects placed near an edge may be cut; use contain with padding for those.
Validation checklist
- Open a few outputs and confirm width, height, and format match the settings.
- Compare before and after bytes in the list for any file that barely shrank.
- Zoom to 100% on screenshots and text-heavy images to check compression artefacts.
- Check for file-level errors before downloading the ZIP so nothing is silently missing.
- Confirm transparent areas are still transparent, or padded with the intended colour.