Resize by width, scale, or box
Add one image or a batch, choose a resize mode, then download individual outputs or a ZIP. The resizer works with JPG, PNG, and WebP files and keeps the source format for each output.
Resizing changes pixel dimensions. It does not choose a new compression format, remove the need for cropping, or guarantee a smaller file in every case. For web images, resize first, then compress if the file is still too large.
Fixed widths
Fixed widths create one or more versions based on target width. Enter values such as 1600,1200,800, and the height is calculated from the original aspect ratio.
This mode fits responsive images because the same source can become several files for srcset, CMS uploads, article images, or gallery thumbnails:
hero-1600x900.jpghero-1200x675.jpghero-800x450.jpg
Scale percentages
Scale mode uses the original dimensions as the reference. Enter 50 for a half-size copy, or 200,100,50 when you need multiple proportional versions from the same source.
Use this mode when relative size matters more than a final pixel width. It also gives you side-by-side sizes from the same source before you choose a publishing size.
Fit inside and cover
Fit inside (W x H) keeps the whole image visible inside a maximum box. A 1200 x 800 box might produce 1200 x 675 for a landscape image or 533 x 800 for a portrait image. No content is cropped.
Cover (W x H) fills the exact target size. If the source ratio does not match, the image is center-cropped. A 1080 x 1080 cover output is always square, even when the original is landscape or portrait.
Use Fit inside for product photos, documentation screenshots, and uploads where every edge must remain visible. Use Cover for cards, avatars, thumbnails, and fixed-size slots where matching dimensions matters more than preserving the full frame.
Batch output
Each image card lists generated files after processing. Output filenames include dimensions, which makes the files easier to audit before upload.
Download all as ZIP processes the batch and groups the generated outputs into an archive. Use smaller batches if the originals are very large or the tab starts using too much memory.
Resize before compression
A camera photo can be 4000px to 8000px wide even when a website only displays it around 1200px. Resizing removes pixels the final layout will never show.
Compression works on the remaining pixels. If a file still needs to be smaller after resizing, pass the resized output through an image compressor or format converter.
Transparency and metadata
PNG and WebP outputs can keep transparent pixels when the source format supports them. JPEG cannot store transparency, so any alpha channel has to be flattened by the browser encoder.
The resize step re-renders pixels through canvas. EXIF, GPS data, camera model, orientation tags, and embedded profiles are commonly stripped from the output.
Exact-size variants
Some related pages open the same resizer with a fixed target size, such as a square thumbnail or social preview dimension. Use those pages when the final dimensions are already known. Use the full resizer when a batch needs mixed widths, percentage scales, or a max-size box.