Batch Image Resizer & Converter
Drop up to 50 photos, pick a resize rule and output format, then download everything as a single ZIP. All processing happens inside your browser, nothing is uploaded anywhere.
Drop images here
JPEG, PNG, WebP, GIF, AVIF · up to 50 files
Resize rule
Max dimension (px)
Output format
Quality ?
Filename convention
Drop or browse images above to get started.
Learn more: fit modes, size estimates and batch output
How the size estimate is worked out
The "Est. output" figure in the stats bar comes from arithmetic on numbers already on screen, without re-encoding anything. The tool scales each source file's size by how much the pixel area shrinks, then multiplies by the quality setting as a fraction.
A 6 MB photo at 4000x3000 resized to 2000x1500 loses three quarters of its pixels, so the estimate starts at 1.5 MB. At quality 85 as a JPEG, it lands near 1.28 MB.
PNG skips the quality step because it is lossless, so a PNG estimate is only the area ratio applied to the source size. JPEG and WebP share the same quality fraction, which means their estimates match at the same slider position even though the real files will not.
Contain, cover and stretch
Exact dimensions ask for a width and a height, and the fit setting decides what happens when your image is a different shape than that box. Contain takes the smaller of the two scale factors, cover takes the larger, and stretch ignores both.
Feed a 4000x2000 panorama into an 800x600 box and contain returns 800x400, which sits inside the box with vertical room left over. Cover returns 1200x600, wider than the box but the same height.
Neither mode adds padding or trims anything. Both keep the original proportions, so the output only matches the numbers you typed when the two aspect ratios already agree. Stretch always returns exactly 800x600, and it gets there by distorting the picture.
Max dimension, longest edge and percentage
Max dimension and longest edge both measure the longer of the two sides. If that side is already at or below your target, the file passes through at its original size, so a 1920 px cap does nothing to a batch of 1200 px images.
Percentage scaling accepts 1 to 500, and anything above 100 enlarges the image. Exact dimensions can also scale up when the box you type is larger than the source, while max dimension and longest edge only ever shrink. Upscaling interpolates new pixels, so the result looks soft.
The quality slider covers 60 to 100 and starts at 85. It disappears when you switch to PNG, since PNG output ignores it.
Formats, filenames and the ZIP
Google's WebP documentation puts lossy WebP at "25-34% smaller than comparable JPEG images at equivalent SSIM quality index". The estimate cannot show you that gap, but each thumbnail badge switches to the real byte count once processing finishes, in green if the file shrank.
Picking Original re-encodes each file as the type it already was, falling back to JPEG for anything that is not JPEG, PNG or WebP. Keeping the original filename still swaps the extension to match the format you exported.
Names can stay as they are, gain a _resized suffix, or become sequential numbers padded to four digits, as in image_0001.jpg. The finished batch downloads as one ZIP called bentoutils-batch followed by the date, assembled in the browser like everything else here.
FAQ
Can I resize 50 photos at once without uploading them?
Yes. A batch holds up to 50 images and they run one after another inside the tab, with the process button showing how far along it is. Nothing is sent to a server, and the finished files come back as a single ZIP.
Why does cover mode not crop my photos to the exact size?
Cover scales the image until it fully covers the width and height you typed, then keeps the whole frame instead of trimming the overflow. The result is at least as large as your box on both sides and keeps the original proportions, so stretch is the only fit that returns your exact numbers.
Why is my exported file a different size than the estimate?
The estimate only knows the pixel area and the quality setting, so it cannot account for what is in the picture. A flat sky compresses much further than dense foliage at the same resolution and quality, which is why the badge updates to the real size after processing.