This model requires you to accept the Apache 2.0 license terms before download.
Description
Swin2SR RealWorld x4 - Swin Transformer V2 super-resolution trained on BSRGAN-style degradations, for real-world photographs carrying noise, blur and compression artifacts rather than clean sources. PSNR-oriented variant, so it favours fidelity over invented detail. Upscales 4x. Apache 2.0.
Tags
At a glance
| Kind | image upscaler |
|---|---|
| Precision | float32 |
| Container size | 46.9 MiB |
| Weights inside | 46.1 MiB |
| Components | 1 |
| Licence | Apache 2.0 |
| Origin | caidas/swin2SR-realworld-sr-x4-64-bsrgan-psnr @ bb13f02e45e8 |
What is inside the container
A .nbx container packs every part of swin2SR-realworld-sr-x4-64-bsrgan-psnr — its single component with weights, graphs and runtime settings — so one file is all the runtime needs.
| Component | Blocks | Layers | Weights |
|---|---|---|---|
| model | 0 | — | 46.1 MiB |
Memory needed
The weights of swin2SR-realworld-sr-x4-64-bsrgan-psnr take 46.1 MiB in float32. That is the floor for the memory the runtime has to place, spread over the devices it finds; activations come on top and grow with the size and length you ask for. A measured peak per device class is not published for this model yet.
Verify what you downloaded
Every download carries its SHA-256, and the file endpoint sends it in the X-NBX-SHA256 header. The container is 46.9 MiB (49,172,114 bytes).
$ sha256sum swin2SR-realworld-sr-x4-64-bsrgan-psnr.nbx
4d535106bba91c2e2322c91983da9a19aa6a8b4b5def7a6740fd52a915ba78f9
Questions
What is swin2SR-realworld-sr-x4-64-bsrgan-psnr?
A image upscaler by caidas, distributed as a float32 .nbx container and licensed Apache 2.0.
How big is the download?
46.9 MiB, of which 46.1 MiB are weights.
How do I run it?
Install the runtime from the download page, then use the commands under “CLI Usage” below: neurobrix import caidas/swin2SR-realworld-sr-x4-64-bsrgan-psnr, then neurobrix serve --model swin2SR-realworld-sr-x4-64-bsrgan-psnr.
CLI Usage
$ neurobrix import caidas/swin2SR-realworld-sr-x4-64-bsrgan-psnr
$ neurobrix serve --model swin2SR-realworld-sr-x4-64-bsrgan-psnr
$ neurobrix run --prompt "your prompt here"
$ neurobrix stop