Resize WebP images by pixels or percentage. Preserves transparency. Free.
WebP is Google's modern image format — smaller than JPG and PNG at the same quality, with full transparency support. Resizing is one of the most common edits, especially for photos and design assets bound for the web. Resize first, compress second — that's the order that gives the smallest, sharpest result. Common reasons to resize:
srcset — Generate multiple sizes (320, 640, 1024, 1920) of one source image to serve via <img srcset>. Each viewport gets the right size, no over-downloading on phones.| Mode | What it does | Best for |
|---|---|---|
| Width × height (px) | Exact target size | Pixel-perfect specs (Instagram 1080×1080, ad slots) |
| Percentage | Proportional scaling | Scale one image to "50% of original" |
| Resolution preset (1080p / 720p / 480p) | Common breakpoints | Quick scaling to standard sizes |
| Original (keep) | No resize | Use just for re-encoding / changing compression |
| Use case | Width × height |
|---|---|
| Mobile thumbnail | 320 × auto |
| Mobile hero | 640 × auto |
| Tablet hero | 1024 × auto |
| Desktop hero | 1920 × auto |
| 4K display | 3840 × auto |
| Instagram square | 1080 × 1080 |
| Instagram Stories | 1080 × 1920 |
| YouTube thumbnail | 1280 × 720 |
| Profile picture | 400 × 400 |
Downscaling (making the image smaller) preserves quality very well — it's essentially throwing away pixels you didn't need. Upscaling (making it larger) can soften the image because the resizer has to invent pixels by interpolation. For natural-looking upscales beyond 2×, an AI upscaler is better than this tool. For all downscaling and modest upscaling (up to 2×), standard resampling produces clean results.
Yes. WebP supports a full 8-bit alpha channel and the resize tool preserves it exactly — logos, icons, and PNG-style transparent assets stay transparent in the output WebP.
Lossless for graphics, screenshots, line art, logos, icons — perfect quality, often smaller than PNG. Lossy for photographs at quality 75-85 — typically 25-50% smaller than lossless WebP and visually indistinguishable. WebP's lossy mode is high quality even at quality 70-75; aggressive settings (50 and below) can show subtle softness on natural skin tones and gradients.
Yes — animated WebPs resize frame-by-frame, with the same dimensions applied to every frame. The output is a smaller animated WebP. If you need a single still frame, see WebP to PNG.
Roughly yes, but not exactly proportional. Halving width × height = 1/4 the pixel area, which usually translates to ~70-80% smaller file (compression is more efficient at smaller sizes for natural images). Halving only one dimension cuts pixel count in half and file size by ~40-60%. Cropping is more aggressive than resizing for shrinking — see Crop WebP.
Not on this page — the editor holds one image at a time; resize it, save, then + Start New for the next file. For a real batch — the same target dimensions applied to a whole image library in one session — use Compress WebP, which runs server-side and includes a resolution step alongside compression.
Resize during conversion if both operations are needed — the conversion step has to re-encode anyway, so combining is more efficient. Examples: PNG to WebP lets you resize on the way out. If your source is already WebP, this resize tool is the right place.
For screen / web: 72 or 96 DPI. For inkjet draft prints: 150 DPI. For high-quality offset printing: 300 DPI. For fine-art prints: 600+ DPI. Note: DPI is metadata that tells the receiving software how big the image should print — it doesn't change the pixel dimensions. To actually change resolution, use the width × height fields.