Add paper link, pipeline tag and sample usage

README.md

@@ -1,25 +1,27 @@
 ---
+base_model:
+- Wan-AI/Wan2.1-T2V-14B
+- Wan-AI/Wan2.1-I2V-14B-480P
+- Wan-AI/Wan2.1-I2V-14B-720P
+library_name: diffusers
 license: apache-2.0
+pipeline_tag: text-to-video
 tags:
 - diffusion-single-file
 - comfyui
 - distillation
 - lora
 - video
-- video genration
-base_model:
-- Wan-AI/Wan2.1-T2V-14B
-- Wan-AI/Wan2.1-I2V-14B-480P
-- Wan-AI/Wan2.1-I2V-14B-720P
-library_name: diffusers
+- video generation
 ---
+
 <div align="center">
 
 # 🎬 Wan2.1 Distilled Models
 
 ### ⚡ High-Performance Video Generation with 4-Step Inference
 
-*Distillation-accelerated versions of Wan2.1 - Dramatically faster while maintaining exceptional quality*
+*Distillation-accelerated versions of Wan2.1 based on the paper [SGMD: Score Gradient Matching Distillation for Few-Step Video Diffusion Distillation](https://huggingface.co/papers/2605.30116) - Dramatically faster while maintaining exceptional quality*
 
 ![image/png](https://cdn-uploads.huggingface.co/production/uploads/680de13385293771bc57400b/gXhUuWyuJpxOwGf5GQ49r.png)
 
@@ -128,7 +130,49 @@ wan2.1_t2v_14b_scaled_fp8_e4m3_lightx2v_4step_comfyui.safetensors  # T2V - FP8 C
 
 **LightX2V is a high-performance inference framework optimized for these models, approximately 2x faster than ComfyUI with better quantization accuracy. Highly recommended!**
 
-#### Quick Start
+### Python Sample Usage
+
+```python
+from lightx2v import LightX2VPipeline
+
+# Initialize pipeline for Wan2.1 I2V task
+pipe = LightX2VPipeline(
+    model_path="lightx2v/Wan2.1-Distill-Models",
+    model_cls="wan2.1",
+    task="i2v",
+)
+
+# Enable offloading to reduce VRAM usage (suitable for consumer GPUs)
+pipe.enable_offload(
+    cpu_offload=True,
+    offload_granularity="block",
+    text_encoder_offload=True,
+    image_encoder_offload=False,
+    vae_offload=False,
+)
+
+# Create generator with 4-step distilled inference
+pipe.create_generator(
+    attn_mode="sage_attn2",
+    infer_steps=4,
+    height=480,
+    width=832,
+    num_frames=81,
+    guidance_scale=5.0,
+    sample_shift=5.0,
+)
+
+# Generate video
+pipe.generate(
+    seed=42,
+    image_path="path/to/image.jpg",
+    prompt="Summer beach vacation style, a white cat wearing sunglasses sits on a surfboard.",
+    negative_prompt="shaking camera, low quality, static",
+    save_result_path="output.mp4",
+)
+```
+
+#### Quick Start (CLI)
 
 1. Download model (720P I2V FP8 example)
 ```bash
@@ -165,7 +209,7 @@ Choose the appropriate configuration based on your GPU memory:
 
 
 5. Run inference
-```
+```bash
 cd scripts
 bash wan/run_wan_i2v_distill_4step_cfg.sh
 ```