Image Resizer
Resize by pixels or percentage with batch support.
Resize photos to exact pixel dimensions or scale them by percentage. Process one image or a whole batch at once — every file stays in your browser the entire time.
About the Image Resizer
Image resizing is one of those tasks that sounds trivial until you actually have to do it well. Anyone can stretch an image to a new size, but doing it without producing soft edges, color fringes, or moiré patterns requires a resampling algorithm and a clear idea of what the output is for. Our resizer uses the browser's native canvas pipeline with high-quality smoothing enabled, which produces clean downscales for nearly every kind of photographic content — portraits, landscapes, product shots, and screenshots alike.
The two modes you'll use most are pixel-based and percentage-based resizing. Pixel-based resizing is what you want when a destination has a strict requirement: a 1080×1080 Instagram square, a 1200×630 Open Graph image, a 32×32 icon, or a 1920×1080 desktop wallpaper. Percentage-based resizing is what you want when you have a folder of photos that are all too big and you simply want everything 50% smaller — uniform reductions are the fastest way to take a batch of 24-megapixel DSLR exports and turn them into something email-friendly without thinking about each one individually.
Aspect ratio is the other piece worth understanding. When the aspect-ratio lock is on, changing one dimension automatically updates the other so the picture keeps its original proportions. Turn the lock off only when you genuinely need a non-proportional output — typically when you're matching a fixed target like a banner slot — and be aware that the result will look stretched. There is no resampling algorithm that can hide a non-proportional resize; the squashing is a property of the geometry, not the math.
Resizing is not the same as compressing. A resized JPG is smaller because it has fewer pixels, but each remaining pixel is still encoded at the same JPG quality the source used. If you want a smaller file at the same dimensions, use the compressor instead. If you want both — fewer pixels and a tighter encoding — resize first, then run the result through the compressor. Doing it in that order gives you the cleanest output because the compressor has fewer pixels to spend its quality budget on.
Everything happens locally. The browser reads the file with the File API, decodes it into an off-screen canvas, scales the canvas into a new size, and re-encodes the result. There's no upload, no temporary file on a server, no telemetry. When you close the tab, every byte of your image is gone from this machine's RAM. That property matters more than people realize: photos contain location data, faces, and context that you may not want trickling out to a third party's analytics pipeline.
How to use the Image Resizer
- 1
Add your images
Drag photos onto the drop zone or click to browse. JPG, PNG, and WebP are all supported, and you can queue as many files as your device can comfortably hold in memory.
- 2
Pick a resize mode
Choose pixel dimensions when you need an exact width or height (for example, 1080×1080 for Instagram). Choose percentage when you want to shrink everything uniformly (for example, 50% of original).
- 3
Lock the aspect ratio
Leave the aspect-ratio lock on to preserve proportions automatically. Turn it off if you need a non-proportional result.
- 4
Download the result
Click Resize, then download each file individually or grab them all. The originals never leave your device — there is nothing to clean up server-side.
Features
- Resize by exact pixel dimensions (width × height)
- Resize by percentage of the original
- Optional aspect-ratio lock
- Batch processing for multiple files at once
- Hover-to-compare original vs. resized preview
- Works fully offline after the first page load
Common use cases
- Shrink high-megapixel phone photos before emailing them
- Prepare images at exact dimensions for Instagram, X, or LinkedIn
- Generate web-ready thumbnails from camera RAW exports
- Bulk-resize product photos to a consistent storefront size
- Create multiple sized versions of an asset for responsive web design
- Fit images under a CMS or forum upload limit without losing visible quality
Tips and best practices
Resize before compressing
If you want both smaller dimensions and a smaller file, always resize first and then compress the smaller image. Compression algorithms perform better when they have fewer pixels to allocate quality across.
Downscale, don't upscale
Going from 4000×3000 to 1200×900 looks great. Going the other direction just enlarges existing pixels and inevitably looks softer. If you need a higher-resolution version, start from the highest-resolution original you have.
Match the platform's actual display size
Instagram displays feed images at 1080 pixels wide. There is no point in uploading a 4000-pixel image — Instagram will just resize it for you, often with worse results than doing it yourself first.
Use percentages for batches
When you're processing a folder of mixed-aspect photos (some landscape, some portrait), percentage mode keeps every photo in proportion. Pixel mode would force everything to the same dimensions, which is rarely what you want.
Keep the original
Save your downscaled version with a new filename. The resizer doesn't overwrite anything, but it's still a good habit so you can always come back to the original if you need a different size later.
Technical details
Common target sizes
Instagram square: 1080×1080. Instagram portrait: 1080×1350. Instagram Stories / Reels: 1080×1920. YouTube thumbnail: 1280×720. Twitter/X header: 1500×500. LinkedIn cover: 1584×396. Open Graph image: 1200×630. Apple Touch icon: 180×180. Favicon: 32×32 or 16×16.
How the resampling works
Downscaling uses the browser's built-in high-quality smoothing, which is typically a bicubic or Lanczos-style filter depending on the engine. Both produce sharp results that avoid the staircase aliasing you get from a naive nearest-neighbor resize. Quality is comparable to desktop tools like Photoshop's Bicubic Sharper for downscales.
Format choices for the output
The output keeps the same format as the source. If you want a different format (for example, JPG output from a PNG source), use the format converter on the result afterwards. Pairing resize → convert → compress in that order gives you the most control over the final file.
Frequently asked questions
Does resizing reduce image quality?
Downscaling (making images smaller) preserves quality well because you are discarding pixels. Upscaling (making them larger) is bounded by the original resolution — you cannot invent detail that was not there, so the result will look softer than the original.
Can I resize hundreds of images at once?
Yes, but memory is the practical limit. A modern laptop handles 20–50 photos at typical phone resolutions without trouble. For very large batches, process them in groups so the browser doesn't run low on memory.
Will EXIF metadata be preserved?
The resize tool focuses on pixel data and does not preserve full EXIF. If you need to keep metadata, use the EXIF Viewer tool first to inspect what you have, then resize. If you specifically want to strip EXIF (for privacy reasons), resizing already does that for you as a side effect.
Why does my resized image look slightly different in color?
If the source image has an embedded ICC color profile, the browser canvas may convert it to sRGB during the resize. For most web use this is exactly what you want, since sRGB is the standard for screens. If you need exact color management for print, do the resize in a desktop tool that respects ICC profiles end-to-end.
What's the difference between resize and crop?
Resize keeps the entire image but changes its dimensions. Crop keeps the dimensions of a selected region but discards everything outside that region. If you need both — for example, a square crop at 1080×1080 — crop first to get the shape, then resize to the target pixel size.
Can I resize to a maximum dimension while keeping aspect ratio?
Yes. With the aspect-ratio lock on, set either the width or the height to your maximum and the other dimension will be calculated automatically. For example, setting width = 1200 on a 4000×3000 source produces 1200×900 — exactly what you'd want for a max-1200 web image.
Is there a maximum file size I can resize?
There's no hard limit imposed by the tool, but very large files (over 50 MB or images larger than ~80 megapixels) may exceed your browser's memory limits. If you hit that wall, try processing one image at a time, or downscale by percentage first to bring things into a manageable range.