add usp for wanx

examples/wanvideo/README.md

@@ -49,6 +49,20 @@ We present a detailed table here. The model is tested on a single A100.
 
 https://github.com/user-attachments/assets/3908bc64-d451-485a-8b61-28f6d32dd92f
 
+### Parallel Inference
+
+1. Unified Sequence Parallel (USP)
+
+```bash
+pip install xfuser>=0.4.3
+```
+
+```bash
+torchrun --standalone --nproc_per_node=8 ./wan_14b_text_to_video_usp.py
+```
+
+2. Tensor Parallel
+
 Tensor parallel module of Wan-Video-14B-T2V is still under development. An example script is provided in [`./wan_14b_text_to_video_tensor_parallel.py`](./wan_14b_text_to_video_tensor_parallel.py).
 
 ### Wan-Video-14B-I2V

examples/wanvideo/wan_14b_text_to_video.py

@@ -1,7 +1,6 @@
 import torch
 from diffsynth import ModelManager, WanVideoPipeline, save_video, VideoData
 from modelscope import snapshot_download
-import torch.distributed as dist
 
 
 # Download models
@@ -24,27 +23,7 @@ model_manager.load_models(
     ],
     torch_dtype=torch.float8_e4m3fn, # You can set `torch_dtype=torch.bfloat16` to disable FP8 quantization.
 )
-
-dist.init_process_group(
-    backend="nccl",
-    init_method="env://",
-)
-from xfuser.core.distributed import (initialize_model_parallel,
-                                     init_distributed_environment)
-init_distributed_environment(
-    rank=dist.get_rank(), world_size=dist.get_world_size())
-
-initialize_model_parallel(
-    sequence_parallel_degree=dist.get_world_size(),
-    ring_degree=1,
-    ulysses_degree=dist.get_world_size(),
-)
-torch.cuda.set_device(dist.get_rank())
-
-pipe = WanVideoPipeline.from_model_manager(model_manager, 
-                                           torch_dtype=torch.bfloat16, 
-                                           device=f"cuda:{dist.get_rank()}", 
-                                           use_usp=True if dist.get_world_size() > 1 else False)
+pipe = WanVideoPipeline.from_model_manager(model_manager, torch_dtype=torch.bfloat16, device="cuda")
 pipe.enable_vram_management(num_persistent_param_in_dit=None) # You can set `num_persistent_param_in_dit` to a small number to reduce VRAM required.
 
 # Text-to-video
@@ -54,4 +33,4 @@ video = pipe(
     num_inference_steps=50,
     seed=0, tiled=True
 )
-save_video(video, "video1.mp4", fps=25, quality=5)
+save_video(video, "video1.mp4", fps=25, quality=5)

examples/wanvideo/wan_14b_text_to_video_usp.py

@@ -0,0 +1,57 @@
+import torch
+from diffsynth import ModelManager, WanVideoPipeline, save_video, VideoData
+from modelscope import snapshot_download
+import torch.distributed as dist
+
+
+# Download models
+snapshot_download("Wan-AI/Wan2.1-T2V-14B", local_dir="models/Wan-AI/Wan2.1-T2V-14B")
+
+# Load models
+model_manager = ModelManager(device="cpu")
+model_manager.load_models(
+    [
+        [
+            "models/Wan-AI/Wan2.1-T2V-14B/diffusion_pytorch_model-00001-of-00006.safetensors",
+            "models/Wan-AI/Wan2.1-T2V-14B/diffusion_pytorch_model-00002-of-00006.safetensors",
+            "models/Wan-AI/Wan2.1-T2V-14B/diffusion_pytorch_model-00003-of-00006.safetensors",
+            "models/Wan-AI/Wan2.1-T2V-14B/diffusion_pytorch_model-00004-of-00006.safetensors",
+            "models/Wan-AI/Wan2.1-T2V-14B/diffusion_pytorch_model-00005-of-00006.safetensors",
+            "models/Wan-AI/Wan2.1-T2V-14B/diffusion_pytorch_model-00006-of-00006.safetensors",
+        ],
+        "models/Wan-AI/Wan2.1-T2V-14B/models_t5_umt5-xxl-enc-bf16.pth",
+        "models/Wan-AI/Wan2.1-T2V-14B/Wan2.1_VAE.pth",
+    ],
+    torch_dtype=torch.float8_e4m3fn, # You can set `torch_dtype=torch.bfloat16` to disable FP8 quantization.
+)
+
+dist.init_process_group(
+    backend="nccl",
+    init_method="env://",
+)
+from xfuser.core.distributed import (initialize_model_parallel,
+                                     init_distributed_environment)
+init_distributed_environment(
+    rank=dist.get_rank(), world_size=dist.get_world_size())
+
+initialize_model_parallel(
+    sequence_parallel_degree=dist.get_world_size(),
+    ring_degree=1,
+    ulysses_degree=dist.get_world_size(),
+)
+torch.cuda.set_device(dist.get_rank())
+
+pipe = WanVideoPipeline.from_model_manager(model_manager, 
+                                           torch_dtype=torch.bfloat16, 
+                                           device=f"cuda:{dist.get_rank()}", 
+                                           use_usp=True if dist.get_world_size() > 1 else False)
+pipe.enable_vram_management(num_persistent_param_in_dit=None) # You can set `num_persistent_param_in_dit` to a small number to reduce VRAM required.
+
+# Text-to-video
+video = pipe(
+    prompt="一名宇航员身穿太空服，面朝镜头骑着一匹机械马在火星表面驰骋。红色的荒凉地表延伸至远方，点缀着巨大的陨石坑和奇特的岩石结构。机械马的步伐稳健，扬起微弱的尘埃，展现出未来科技与原始探索的完美结合。宇航员手持操控装置，目光坚定，仿佛正在开辟人类的新疆域。背景是深邃的宇宙和蔚蓝的地球，画面既科幻又充满希望，让人不禁畅想未来的星际生活。",
+    negative_prompt="色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走",
+    num_inference_steps=50,
+    seed=0, tiled=True
+)
+save_video(video, "video1.mp4", fps=25, quality=5)

requirements.txt

@@ -11,4 +11,3 @@ sentencepiece
 protobuf
 modelscope
 ftfy
-xfuser>=0.4.2