Upscale images with AI
Runs Real-ESRGAN inside your browser to enlarge small or blurry images 2× or 4×. It reconstructs edges and texture, so the result is far cleaner than simple stretching. Images never leave your device.
The first run downloads an AI model (about 64 MB); later runs use the copy stored in your browser. Your files are never uploaded.
Settings
How to use
- 1
Add images
Drop the images you want to enlarge. Works well for small photos, screenshots and old images.
- 2
Choose the scale
2× or 4×. The first run downloads the AI model (about 64 MB). With WebGPU it takes seconds; without it, tens of seconds to minutes.
- 3
Run and check
Use “Compare” to check against the original. After processing you see whether it ran on the GPU or CPU.
When to use 2× or 4×, and which images suit upscaling, are explained in the article “How to Upscale Images with AI” on our blog.
How it works, strengths and limits
Real-ESRGAN is a super-resolution model that reconstructs plausible detail. It suits photo texture, illustration line work and small images saved from the web. It cannot recover text or facial detail that downscaling erased completely, and faces tend to come out slightly smooth.
With WebGPU it takes seconds; without it, the CPU (WebAssembly) path takes tens of seconds to minutes. Very large images are reduced first and you are told so. The model (about 64 MB) is downloaded once and read from the browser cache afterwards.
Frequently asked questions
Are my images sent to a server?
No. The AI model is downloaded into your browser and inference runs on your device, so images are never uploaded. Processing statistics used to fix bugs (device specs, error types and so on; never the images themselves or their file names) are sent to this site's server. You can turn this off with the switch under “Help us improve the service” at the bottom of the page.
How long does it take?
In Chrome, Edge or Safari with WebGPU, a 1000×1000 image at 2× takes a few seconds. Without WebGPU the same image can take 1–3 minutes. Afterwards you see whether it ran on the GPU or CPU.
Can I upscale large images?
To protect device memory there is a cap on input pixels (about 3 megapixels with WebGPU, 0.48 with CPU). Larger images are reduced first and you are told so.
Why do faces look too smooth?
That is a known trait of Real-ESRGAN. To sharpen faces, use the “Face restoration” tool (GFPGAN); with its “Upscale first” option you get enlargement and face restoration in one pass. For whole photos this tool's quality is more than adequate.
Help us improve the service
If a conversion fails, sending us the circumstances helps us track down the cause. Knowing which devices and formats fail most often, and how long processing takes, lets us revise the supported formats and fix the bugs behind them. If you don't mind, leaving this on is a real help.
What is sent
- The tool you used and the settings you chose (output format, quality, etc.)
- Your device specs (GPU, memory, core count)
- The input format with a rough size
- How many files succeeded or failed and how long it took
- The error type and details when something fails
What is never sent
- The files themselves and file names
- Text, subtitles or prompts you add to a video
- The text to be read aloud, voice descriptions, the pronunciation dictionary and saved voices
- The text in PDFs, the text you rewrote or redacted, passwords, font names and the page ranges you enter
- IP addresses
The data goes only to this site's own server, never to a third-party service.
Details sent by specific tools (counts, times, kinds and yes/no only)
In addition to “What is sent”, text to speech and the PDF tools send the following details.
- Text to speech
- The approximate number of characters and sentences, how many parts the text was generated in, and how long the AI model took to load.
- PDF tools (all)
- The page count range, whether editing is restricted, whether a permissions password was entered, and whether the PDF is signed or encrypted.
- PDF text editing and redaction
- The numbers of rewrites, deletions, white-outs and redactions, the number of rewritten characters by source (original font or fallback font, the latter as a range) and the fallback typeface, the kinds of reasons why lines cannot be edited, and the number of pages turned into images by redaction and the reasons.
- PDF page operations and conversion between images and PDF
- The numbers of inputs and outputs, of pages rotated, deleted, duplicated or moved, of images inserted as they are, converted, in HEIC format or with corrected orientation, and of pages rendered at a lower resolution (all as ranges), whether the original was updated in place or a new document was created, and whether the original protection was reapplied, form fields were written into the page and bookmarks were added back.
- PDF compression
- The numbers of inputs and outputs, of images downsized, not downsized and black-and-white, of downsized images whose transparency was also downsized, that stayed lossless or that were converted from CMYK to RGB, and how many times the file was compressed (all as ranges), the kinds of reasons images were not downsized, whether the target size was met and the kind of reason it was not met, the range of how much smaller the file got, whether the original protection was reapplied, and whether the original PDF was returned unchanged.
- PDF cropping and page numbers
- The numbers of inputs and outputs and of pages cropped, numbered or with their margins reduced (all as ranges), how the margins were set, how the pages to crop or number were chosen, the number format and position, the range of the starting number, that the original was updated in place, and whether the original protection was reapplied.