Update README.md

README.md

@@ -1,26 +1,23 @@
 ---
 license: other
-license_name: attribution-noncommercial-4.0-international
-license_link: LICENSE
+license_name: flux-1-dev-non-commercial-license
 tags:
-- text
-- images
 - text-to-image
+- SVDQuant
+- FLUX.1-dev
+- INT4
+- FLUX.1
+- Diffusion
+- Quantization
+- LoRA
 language:
 - en
-source_datasets:
-- sDCI
-task_categories:
-- text-to-image
-dataset_info:
-  features:
-    - name: filename
-      dtype: string
-    - name: image
-      dtype: image
-    - name: prompt
-      dtype: string
-arxiv: 2411.05007
+base_model:
+- mit-han-lab/svdq-int4-flux.1-dev
+pipeline_tag: text-to-image
+datasets:
+- mit-han-lab/svdquant-datasets
+library_name: diffusers
 ---
 
 <p align="center" style="border-radius: 10px">
@@ -34,9 +31,54 @@ arxiv: 2411.05007
   <a href='https://hanlab.mit.edu/projects/svdquant'>[Website]</a>&ensp;
   <a href='https://hanlab.mit.edu/blog/svdquant'>[Blog]</a>
 </div>
-This is the [sDCI](https://arxiv.org/abs/2312.08578) dataset used in [SVDQuant](https://hanlab.mit.edu/blog/svdquant) for benchmarking. 
+![teaser](https://github.com/mit-han-lab/nunchaku/raw/refs/heads/main/assets/lora.jpg)
+SVDQuant seamlessly integrates with off-the-shelf LoRAs without requiring re-quantization. When applying LoRAs, it matches the image quality of the original 16-bit FLUX.1-dev.
+
+## Model Description
+
+<div>
+This reposity contains a converted LoRA collection for SVDQuant INT4 FLUX.1-dev. The LoRA style includes <a href="https://huggingface.co/XLabs-AI/flux-RealismLora">Realism</a>, 
+<a href="https://huggingface.co/aleksa-codes/flux-ghibsky-illustration">Ghibsky Illustration</a>, 
+<a href="https://huggingface.co/alvdansen/sonny-anime-fixed">Anime</a>, 
+<a href="https://huggingface.co/Shakker-Labs/FLUX.1-dev-LoRA-Children-Simple-Sketch">Children Sketch</a>, and 
+<a href="https://huggingface.co/linoyts/yarn_art_Flux_LoRA">Yarn Art</a>.
+</div>
+
+
+## Usage
+
+### Diffusers
+
+Please follow the instructions in [mit-han-lab/nunchaku](https://github.com/mit-han-lab/nunchaku) to set up the environment. Then you can run the model with
+
+```python
+import torch
+
+from nunchaku.pipelines import flux as nunchaku_flux
+
+pipeline = nunchaku_flux.from_pretrained(
+    "black-forest-labs/FLUX.1-dev",
+    torch_dtype=torch.bfloat16,
+    qmodel_path="mit-han-lab/svdq-int4-flux.1-dev",  # download from Huggingface
+).to("cuda")
+pipeline.transformer.nunchaku_update_params(mit-han-lab/svdquant-models/svdq-flux.1-dev-lora-anime.safetensors)
+pipeline.transformer.nunchaku_set_lora_scale(1)
+image = pipeline("a dog wearing a wizard hat", num_inference_steps=28, guidance_scale=3.5).images[0]
+image.save("example.png")
+```
+
+### Comfy UI
+
+Work in progress.
+
+## Limitations
+
+- The model is only runnable on NVIDIA GPUs with architectures sm_86 (Ampere: RTX 3090, A6000), sm_89 (Ada: RTX 4090), and sm_80 (A100). See this [issue](https://github.com/mit-han-lab/nunchaku/issues/1) for more details.
+- You may observe some slight differences from the BF16 models in details.
+
+### Citation
 
-If you find this dataset useful or relevant to your research, please cite
+If you find this model useful or relevant to your research, please cite
 
 ```bibtex
 @article{
@@ -46,13 +88,4 @@ If you find this dataset useful or relevant to your research, please cite
   journal={arXiv preprint arXiv:2411.05007},
   year={2024}
 }
-
-@inproceedings{urbanek2024picture,
-  title={A picture is worth more than 77 text tokens: Evaluating clip-style models on dense captions},
-  author={Urbanek, Jack and Bordes, Florian and Astolfi, Pietro and Williamson, Mary and Sharma, Vasu and Romero-Soriano, Adriana},
-  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
-  pages={26700--26709},
-  year={2024}
-}
-```
-
+```