| # Image upscaler (browser build) | |
| Unmodified TensorFlow.js layer models from [@upscalerjs/esrgan-slim](https://www.npmjs.com/package/@upscalerjs/esrgan-slim) version 1.0.0 (MIT, see `LICENSE`). | |
| Source: https://github.com/thekevinscott/UpscalerJS | |
| These are the slim ESRGAN 2×, 3× and 4× models WaveSpeed's free image enhancer runs with UpscalerJS. Weights were copied from jsDelivr at that exact package version: | |
| - `esrgan-slim-x2/group1-shard1of1.bin` — 888,300 bytes | |
| - `esrgan-slim-x3/group1-shard1of1.bin` — 907,260 bytes | |
| - `esrgan-slim-x4/group1-shard1of1.bin` — 933,804 bytes | |
| Input `LR`: float32 NHWC, RGB scaled to 0–1, variable height and width. Output is the same layout at 2×, 3× or 4×. | |
Xet Storage Details
- Size:
- 712 Bytes
- Xet hash:
- 24f53b8b589012245b07da735ed194c10abef50f44599c9401d0683eb3c3cedc
·
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.