Update README.md

README.md

@@ -25,7 +25,7 @@ tags:
 
 <div align="center">
 
-[[🤗 <strong><span style="color:hotpink">Diffusers</span></strong> Implementation](https://github.com/huggingface/diffusers/tree/main/src/diffusers/pipelines/visualcloze)]
+[[🤗 <strong><span style="color:hotpink">Diffusers</span></strong> Implementation](https://github.com/huggingface/diffusers/tree/main/src/diffusers/pipelines/visualcloze)] &emsp; [[🤗 LoRA Model Card for Diffusers]](https://huggingface.co/VisualCloze/VisualClozePipeline-LoRA-384)
 
 </div>
 
@@ -41,6 +41,7 @@ If you find VisualCloze is helpful, please consider to star ⭐ the [<strong><sp
 
 ## 📰 News
 - [2025-5-15] 🤗🤗🤗 VisualCloze has been merged into the [<strong><span style="color:hotpink">official pipelines of diffusers</span></strong>](https://github.com/huggingface/diffusers/tree/main/src/diffusers/pipelines/visualcloze).
+- [2025-5-18] 🥳🥳🥳 We have released the LoRA weights supporting diffusers at [LoRA Model Card 384](https://huggingface.co/VisualCloze/VisualClozePipeline-LoRA-384) and [LoRA Model Card 512](https://huggingface.co/VisualCloze/VisualClozePipeline-LoRA-512).
 
 ## 🌠 Key Features
 
@@ -65,9 +66,13 @@ pip install git+https://github.com/huggingface/diffusers.git
 
 [![Huggingface VisualCloze](https://img.shields.io/static/v1?label=Demo&message=Huggingface%20Gradio&color=orange)](https://huggingface.co/spaces/VisualCloze/VisualCloze)
 
-A model trained with the `resolution` of 512 is released at [Model Card](https://huggingface.co/VisualCloze/VisualClozePipeline-512), 
+This model provides the full parameters of our VisualCloze. 
+If you find the download size too large, you can use the [LoRA version](https://huggingface.co/VisualCloze/VisualClozePipeline-LoRA-384) 
+with the FLUX.1-Fill-dev as the base model.
+
+A model trained with the `resolution` of 512 is released at [Full Model Card 512](https://huggingface.co/VisualCloze/VisualClozePipeline-512) and [LoRA Model Card 512](https://huggingface.co/VisualCloze/VisualClozePipeline-LoRA-512), 
 while this model uses the `resolution` of 384. The `resolution` means that each image will be resized to it before being
-concatenated to avoid the out-of-memory error. To generate high-resolution images, we use the [SDEdit](https://arxiv.org/abs/2108.01073) technology for upsampling the generated results.
+concatenated to avoid the out-of-memory error. To generate high-resolution images, we use the SDEdit technology for upsampling the generated results.
 
 #### Example with Depth-to-Image:
 
@@ -106,6 +111,11 @@ high contrast, photorealistic, intimate, elegant, visually balanced, serene atmo
 pipe = VisualClozePipeline.from_pretrained("VisualCloze/VisualClozePipeline-384", resolution=384, torch_dtype=torch.bfloat16)
 pipe.to("cuda")
 
+# Loading the VisualClozePipeline via LoRA
+# pipe = VisualClozePipeline.from_pretrained("black-forest-labs/FLUX.1-Fill-dev", resolution=384, torch_dtype=torch.bfloat16)
+# pipe.load_lora_weights('VisualCloze/VisualClozePipeline-LoRA-384', weight_name='visualcloze-lora-384.safetensors')
+# pipe.to("cuda")
+
 # Run the pipeline
 image_result = pipe(
     task_prompt=task_prompt,
@@ -160,6 +170,11 @@ content_prompt = None
 pipe = VisualClozePipeline.from_pretrained("VisualCloze/VisualClozePipeline-384", resolution=384, torch_dtype=torch.bfloat16)
 pipe.to("cuda")
 
+# Loading the VisualClozePipeline via LoRA
+# pipe = VisualClozePipeline.from_pretrained("black-forest-labs/FLUX.1-Fill-dev", resolution=384, torch_dtype=torch.bfloat16)
+# pipe.load_lora_weights('VisualCloze/VisualClozePipeline-LoRA-384', weight_name='visualcloze-lora-384.safetensors')
+# pipe.to("cuda")
+
 # Run the pipeline
 image_result = pipe(
     task_prompt=task_prompt,