报错
root@xfy:/home# python generate.py --task t2v-1.3B --size 832480 --ckpt_dir models/Wan2.1-T2V-1.3B --prompt "Two anthropomorphic cats in comfy boxing gear and bright gloves fight intensely on a spotlighted stage."
[2025-07-30 15:31:04,466] INFO: offload_model is not specified, set to True.
[2025-07-30 15:31:04,466] INFO: Generation job args: Namespace(task='t2v-1.3B', size='832480', frame_num=81, ckpt_dir='models/Wan2.1-T2V-1.3B', offload_model=True, ulysses_size=1, ring_size=1, t5_fsdp=False, t5_cpu=False, dit_fsdp=False, save_file=None, prompt='Two anthropomorphic cats in comfy boxing gear and bright gloves fight intensely on a spotlighted stage.', use_prompt_extend=False, prompt_extend_method='local_qwen', prompt_extend_model=None, prompt_extend_target_lang='ch', base_seed=8531901343496045993, image=None, sample_solver='unipc', sample_steps=50, sample_shift=5.0, sample_guide_scale=5.0)
[2025-07-30 15:31:04,466] INFO: Generation model config: {'name': 'Config: Wan T2V 1.3B', 't5_model': 'umt5_xxl', 't5_dtype': torch.bfloat16, 'text_len': 512, 'param_dtype': torch.bfloat16, 'num_train_timesteps': 1000, 'sample_fps': 16, 'sample_neg_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', 't5_checkpoint': 'models_t5_umt5-xxl-enc-bf16.pth', 't5_tokenizer': 'google/umt5-xxl', 'vae_checkpoint': 'Wan2.1_VAE.pth', 'vae_stride': (4, 8, 8), 'patch_size': (1, 2, 2), 'dim': 1536, 'ffn_dim': 8960, 'freq_dim': 256, 'num_heads': 12, 'num_layers': 30, 'window_size': (-1, -1), 'qk_norm': True, 'cross_attn_norm': True, 'eps': 1e-06}
[2025-07-30 15:31:04,467] INFO: Input prompt: Two anthropomorphic cats in comfy boxing gear and bright gloves fight intensely on a spotlighted stage.
[2025-07-30 15:31:04,467] INFO: Creating WanT2V pipeline.
[2025-07-30 15:32:35,603] INFO: loading models/Wan2.1-T2V-1.3B/models_t5_umt5-xxl-enc-bf16.pth
[2025-07-30 15:32:51,708] INFO: loading models/Wan2.1-T2V-1.3B/Wan2.1_VAE.pth
[2025-07-30 15:32:52,274] INFO: Creating WanModel from models/Wan2.1-T2V-1.3B
[2025-07-30 15:32:56,472] WARNING: WARNING[XFORMERS]: xFormers can't load C++/CUDA extensions. xFormers was built for:
PyTorch 2.1.0 with CUDA None (you have 2.3.0)
Python 3.10.12 (you have 3.10.12)
Please reinstall xformers (see https://github.com/facebookresearch/xformers#installing-xformers)
Memory-efficient attention, SwiGLU, sparse and more won't be available.
Set XFORMERS_MORE_DETAILS=1 for more details
[2025-07-30 15:33:29,582] INFO: Generating video ...
0%| | 0/50 [00:00<?, ?it/s]
Traceback (most recent call last):
File "/home/generate.py", line 411, in
generate(args)
File "/home/generate.py", line 313, in generate
video = wan_t2v.generate(
File "/home/wan/text2video.py", line 236, in generate
noise_pred_cond = self.model(
File "/usr/local/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/usr/local/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
return forward_call(*args, **kwargs)
File "/home/wan/modules/model.py", line 564, in forward
x = block(x, **kwargs)
File "/usr/local/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/usr/local/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
return forward_call(*args, **kwargs)
File "/home/wan/modules/model.py", line 298, in forward
y = self.self_attn(
File "/usr/local/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/usr/local/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
return forward_call(*args, **kwargs)
File "/home/wan/modules/model.py", line 146, in forward
x = flash_attention(
File "/home/wan/modules/attention.py", line 113, in flash_attention
x = flash_attn.flash_attn_varlen_func(
File "/usr/local/lib/python3.10/site-packages/flash_attn/flash_attn_interface.py", line 1125, in flash_attn_varlen_func
return FlashAttnVarlenFunc.apply(
File "/usr/local/lib/python3.10/site-packages/torch/autograd/function.py", line 598, in apply
return super().apply(*args, **kwargs) # type: ignore[misc]
File "/usr/local/lib/python3.10/site-packages/flash_attn/flash_attn_interface.py", line 621, in forward
out, q, k, v, out_padded, softmax_lse, S_dmask, rng_state = _flash_attn_varlen_forward(
File "/usr/local/lib/python3.10/site-packages/flash_attn/flash_attn_interface.py", line 89, in _flash_attn_varlen_forward
out, q, k, v, out_padded, softmax_lse, S_dmask, rng_state = flash_attn_cuda.varlen_fwd(
RuntimeError: HIP error: invalid device function
HIP kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
For debugging consider passing AMD_SERIALIZE_KERNEL=3.
Compile with TORCH_USE_HIP_DSA to enable device-side assertions.