Image-to-Text
Diffusers
Safetensors
English
uniar
image-generation
image-understanding
image-editing
multimodal
autoregressive
text-to-image
unified-model
Instructions to use ShareLab-SII/UniAR-RL with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use ShareLab-SII/UniAR-RL with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("ShareLab-SII/UniAR-RL", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files
README.md
CHANGED
|
@@ -18,7 +18,7 @@ base_model: ShareLab-SII/UniAR-SFT
|
|
| 18 |
|
| 19 |
**UniAR** is a unified autoregressive multimodal model for **image understanding**, **image generation**, and **image editing** in a single Transformer. UniAR-RL is obtained by reinforcement learning (GRPO) on top of [UniAR-SFT](https://huggingface.co/ShareLab-SII/UniAR-SFT), achieving state-of-the-art text rendering and instruction-following performance among unified models.
|
| 20 |
|
| 21 |
-
[](https://sharelab-sii.github.io/uniar-web)
|
| 23 |
[](https://github.com/ShareLab-SII/UniAR)
|
| 24 |
|
|
|
|
| 18 |
|
| 19 |
**UniAR** is a unified autoregressive multimodal model for **image understanding**, **image generation**, and **image editing** in a single Transformer. UniAR-RL is obtained by reinforcement learning (GRPO) on top of [UniAR-SFT](https://huggingface.co/ShareLab-SII/UniAR-SFT), achieving state-of-the-art text rendering and instruction-following performance among unified models.
|
| 20 |
|
| 21 |
+
[](https://arxiv.org/abs/2606.18249)
|
| 22 |
[](https://sharelab-sii.github.io/uniar-web)
|
| 23 |
[](https://github.com/ShareLab-SII/UniAR)
|
| 24 |
|