LibrePRXPixel7b / README.md
Xuban's picture
Mirror Photoroom/prxpixel-t2i@bcd5e63f0722
9fc4908 verified
|
Raw History Blame Contribute Delete
1.85 kB
---
license: apache-2.0
library_name: libreyolo
pipeline_tag: text-to-image
base_model: Photoroom/prxpixel-t2i
tags:
- text-to-image
- prx
- photoroom
- diffusers
- libreyolo
---
# LibrePRXPixel7b
PRXPixel, Photoroom's 7B pixel-space text-to-image model. It denoises RGB directly, with no autoencoder, and is conditioned on a Qwen3-VL text encoder. Native resolution 1024. Mirrored for LibreYOLO.
```python
from libreyolo import LibreGen
model = LibreGen("prxpixel-7b") # downloads this repo
model("a red fox sleeping in fresh snow").save("fox.png")
```
```bash
libreyolo generate prompt="a red fox" model=prxpixel-7b
```
Install `libreyolo[gen]`. Needs `diffusers>=0.39`.
## This is a mirror, not a new model
The weights are Photoroom's. Nothing was retrained, converted or re-encoded;
every file is byte-identical to upstream and keeps upstream's diffusers
pipeline layout, so `diffusers` loads this folder as it is.
- Upstream repository: [`Photoroom/prxpixel-t2i`](https://huggingface.co/Photoroom/prxpixel-t2i)
- Upstream revision: `bcd5e63f072257a220c5d0ba039c97657398b1c2`
- Announcement: [We're open-sourcing our text-to-image model and the process behind it](https://huggingface.co/blog/Photoroom/prx-open-source-t2i-model)
| File | SHA-256 |
| --- | --- |
| `text_encoder/model.safetensors` | `4d212f35fd3ef50b5c03d3603153ee782df085c8e7286e09fe06d655bd0d2bf3` |
| `transformer/diffusion_pytorch_model-00001-of-00002.safetensors` | `856f4e24d89a7de09a00b8d3e1871406b8f5af7b86b308dc4b7e4e1c418d9a21` |
| `transformer/diffusion_pytorch_model-00002-of-00002.safetensors` | `d05f4a9d276a564bf1c5fc9450352b66179bb5563dcfa71dfc77db67cb4ebcd7` |
## License
Apache-2.0, as published by Photoroom. The bundled text encoder is Alibaba's Qwen3-VL-Embedding-2B (Apache-2.0). See [`LICENSE`](./LICENSE) and [`NOTICE`](./NOTICE).