Cannot find a working triton installation. Either the package is not installed or it is too old.

[15036304416q-create]

ComfyUI Error Report

Error Details

• Node ID: 27
• Node Type: WanVideoSampler
• Exception Type: torch._inductor.exc.TritonMissing
• Exception Message: Cannot find a working triton installation. Either the package is not installed or it is too old. More information on installing Triton can be found at: https://github.com/triton-lang/triton

Set TORCHDYNAMO_VERBOSE=1 for the internal stack trace (please do this especially if you're reporting a bug to PyTorch). For even more developer context, set TORCH_LOGS="+dynamo"

Stack Trace

  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 496, in execute
    output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
                                                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 315, in get_output_data
    return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 289, in _async_map_node_over_list
    await process_inputs(input_dict, i)

  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 277, in process_inputs
    result = f(**inputs)
             ^^^^^^^^^^^

  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-WanVideoWrapper\nodes_sampler.py", line 2906, in process
    raise e

  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-WanVideoWrapper\nodes_sampler.py", line 2796, in process
    noise_pred, self.cache_state = predict_with_cfg(
                                   ^^^^^^^^^^^^^^^^^

  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-WanVideoWrapper\nodes_sampler.py", line 1445, in predict_with_cfg
    raise e

  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-WanVideoWrapper\nodes_sampler.py", line 1318, in predict_with_cfg
    noise_pred_cond, cache_state_cond = transformer(
                                        ^^^^^^^^^^^^

  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1773, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1784, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-WanVideoWrapper\wanvideo\modules\model.py", line 2708, in forward
    x, x_ip, lynx_ref_feature = block(x, x_ip=x_ip, lynx_ref_feature=lynx_ref_feature, **kwargs) #run block
                                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_dynamo\eval_frame.py", line 375, in __call__
    return super().__call__(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1773, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1784, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_dynamo\eval_frame.py", line 749, in compile_wrapper
    raise e.remove_dynamo_frames() from None  # see TORCHDYNAMO_VERBOSE=1
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\scheduler.py", line 4042, in create_backend
    raise TritonMissing(inspect.currentframe())

System Information

• ComfyUI Version: 0.3.64
• Arguments: ComfyUI\main.py --fast fp16_accumulation
• OS: nt
• Python Version: 3.12.10 (tags/v3.12.10:0cc8128, Apr 8 2025, 12:21:36) [MSC v.1943 64 bit (AMD64)]
• Embedded Python: true
• PyTorch Version: 2.8.0+cu128

Devices

• Name: cuda:0 NVIDIA GeForce RTX 4090 : cudaMallocAsync
  • Type: cuda
  • VRAM Total: 25756696576
  • VRAM Free: 24108859392
  • Torch VRAM Total: 0
  • Torch VRAM Free: 0

Logs

2025-10-11T13:35:58.318176 - * CLIPVisionLoader 71:
2025-10-11T13:35:58.322176 -   - Value not in list: clip_name: 'clip_vision_vit_h.safetensors' not in ['CLIP-ViT-H-14-laion2B-s32B-b79K.safetensors.safetensors', 'CLIP-ViT-bigG-14-laion2B-39B-b160k.safetensors.safetensors', 'clip_vision_g.safetensors', 'clip_vision_h.safetensors', 'sigclip_vision_patch14_384.safetensors']
2025-10-11T13:35:58.325178 - Output will be ignored
2025-10-11T13:35:58.329176 - Failed to validate prompt for output 30:
2025-10-11T13:35:58.332177 - Output will be ignored
2025-10-11T13:35:58.335180 - Failed to validate prompt for output 204:
2025-10-11T13:35:58.339180 - Output will be ignored
2025-10-11T13:36:53.901379 - 
Extracting keypoints: 100%|████████████████████████████████████████████████████████████| 97/97 [01:19<00:00,  1.19it/s]2025-10-11T13:36:53.908883 - 
Extracting keypoints: 100%|████████████████████████████████████████████████████████████| 97/97 [01:19<00:00,  1.22it/s]2025-10-11T13:36:53.913883 - 
2025-10-11T13:36:55.729850 - 
Drawing pose images:  72%|███████████████████████████████████████████▎                | 70/97 [00:00<00:00, 346.14it/s]2025-10-11T13:36:55.864502 - 
Drawing pose images: 100%|████████████████████████████████████████████████████████████| 97/97 [00:00<00:00, 287.83it/s]2025-10-11T13:36:55.864502 - 
2025-10-11T13:36:57.247232 - model_path: 2025-10-11T13:36:57.247232 -  2025-10-11T13:36:57.247232 - D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\models\sam2\sam2_hiera_base_plus.safetensors2025-10-11T13:36:57.247232 - 
2025-10-11T13:36:57.248232 - Using model config: D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-segment-anything-2\sam2_configs\sam2_hiera_b+.yaml2025-10-11T13:36:57.249234 - 
2025-10-11T13:36:58.204317 - Resizing to model input image size: 2025-10-11T13:36:58.204317 -  2025-10-11T13:36:58.204317 - 10242025-10-11T13:36:58.204317 - 
2025-10-11T13:36:59.903840 - combined labels: 2025-10-11T13:36:59.903840 -  2025-10-11T13:36:59.903840 - [1. 1. 1. 1. 1. 1. 1. 1.]2025-10-11T13:36:59.903840 - 
2025-10-11T13:36:59.903840 - combined labels shape: 2025-10-11T13:36:59.903840 -  2025-10-11T13:36:59.903840 - (8,)2025-10-11T13:36:59.903840 - 
2025-10-11T13:37:06.466212 - 
propagate in video:  99%|█████████████████████████████████████████████████████████████▎| 96/97 [00:05<00:00, 17.87it/s]2025-10-11T13:37:06.510233 - 
propagate in video: 100%|██████████████████████████████████████████████████████████████| 97/97 [00:05<00:00, 18.76it/s]2025-10-11T13:37:06.510233 - 
2025-10-11T13:37:12.511036 - 
BlockifyMask batch:  92%|███████████████████████████████████████████████████████▉     | 89/97 [00:00<00:00, 144.38it/s]2025-10-11T13:37:12.561543 - 
BlockifyMask batch: 100%|█████████████████████████████████████████████████████████████| 97/97 [00:00<00:00, 143.81it/s]2025-10-11T13:37:12.561543 - 
2025-10-11T13:37:14.792725 - 
DrawMaskOnImage batch:  98%|█████████████████████████████████████████████████████████▊ | 95/97 [00:01<00:00, 50.65it/s]2025-10-11T13:37:14.821795 - 
DrawMaskOnImage batch: 100%|███████████████████████████████████████████████████████████| 97/97 [00:01<00:00, 50.41it/s]2025-10-11T13:37:14.821795 - 
2025-10-11T13:37:18.578804 - Prompt executed in 171.44 seconds
2025-10-11T13:37:19.045444 - Prompt executed in 0.06 seconds
2025-10-11T13:37:47.436117 - got prompt
2025-10-11T13:37:47.459128 - Failed to validate prompt for output 189:
2025-10-11T13:37:47.460128 - * CLIPVisionLoader 71:
2025-10-11T13:37:47.460128 -   - Value not in list: clip_name: 'clip_vision_vit_h.safetensors' not in ['CLIP-ViT-H-14-laion2B-s32B-b79K.safetensors.safetensors', 'CLIP-ViT-bigG-14-laion2B-39B-b160k.safetensors.safetensors', 'clip_vision_g.safetensors', 'clip_vision_h.safetensors', 'sigclip_vision_patch14_384.safetensors']
2025-10-11T13:37:47.460128 - Output will be ignored
2025-10-11T13:37:47.460128 - Failed to validate prompt for output 30:
2025-10-11T13:37:47.460128 - Output will be ignored
2025-10-11T13:37:47.460128 - Failed to validate prompt for output 204:
2025-10-11T13:37:47.460128 - Output will be ignored
2025-10-11T13:37:47.521317 - Prompt executed in 0.06 seconds
2025-10-11T13:37:48.090410 - Exception in callback _ProactorBasePipeTransport._call_connection_lost(None)
handle: <Handle _ProactorBasePipeTransport._call_connection_lost(None)>
Traceback (most recent call last):
  File "asyncio\events.py", line 88, in _run
  File "asyncio\proactor_events.py", line 165, in _call_connection_lost
ConnectionResetError: [WinError 10054] 远程主机强迫关闭了一个现有的连接。
2025-10-11T13:37:57.489800 - �[91mWaiting for SD client to connect...�[0m2025-10-11T13:37:57.489800 - 
2025-10-11T13:43:06.943529 - got prompt
2025-10-11T13:43:07.036732 - Downloading GMMI-VFI model to: D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\models\interpolation\gimm-vfi\gimmvfi_r_arb_lpips_fp32.safetensors
2025-10-11T13:43:12.041549 - !!! Exception during processing !!! An error happened while trying to locate the files on the Hub and we cannot find the appropriate snapshot folder for the specified revision on the local disk. Please check your internet connection and try again.
2025-10-11T13:43:12.046252 - Traceback (most recent call last):
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\urllib3\connectionpool.py", line 773, in urlopen
    self._prepare_proxy(conn)
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\urllib3\connectionpool.py", line 1042, in _prepare_proxy
    conn.connect()
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\urllib3\connection.py", line 790, in connect
    sock_and_verified = _ssl_wrap_socket_and_match_hostname(
                        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\urllib3\connection.py", line 969, in _ssl_wrap_socket_and_match_hostname
    ssl_sock = ssl_wrap_socket(
               ^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\urllib3\util\ssl_.py", line 480, in ssl_wrap_socket
    ssl_sock = _ssl_wrap_socket_impl(sock, context, tls_in_tls, server_hostname)
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\urllib3\util\ssl_.py", line 524, in _ssl_wrap_socket_impl
    return ssl_context.wrap_socket(sock, server_hostname=server_hostname)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "ssl.py", line 455, in wrap_socket
  File "ssl.py", line 1041, in _create
  File "ssl.py", line 1319, in do_handshake
ConnectionResetError: [WinError 10054] 远程主机强迫关闭了一个现有的连接。

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\requests\adapters.py", line 644, in send
    resp = conn.urlopen(
           ^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\urllib3\connectionpool.py", line 841, in urlopen
    retries = retries.increment(
              ^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\urllib3\util\retry.py", line 474, in increment
    raise reraise(type(error), error, _stacktrace)
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\urllib3\util\util.py", line 38, in reraise
    raise value.with_traceback(tb)
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\urllib3\connectionpool.py", line 773, in urlopen
    self._prepare_proxy(conn)
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\urllib3\connectionpool.py", line 1042, in _prepare_proxy
    conn.connect()
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\urllib3\connection.py", line 790, in connect
    sock_and_verified = _ssl_wrap_socket_and_match_hostname(
                        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\urllib3\connection.py", line 969, in _ssl_wrap_socket_and_match_hostname
    ssl_sock = ssl_wrap_socket(
               ^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\urllib3\util\ssl_.py", line 480, in ssl_wrap_socket
    ssl_sock = _ssl_wrap_socket_impl(sock, context, tls_in_tls, server_hostname)
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\urllib3\util\ssl_.py", line 524, in _ssl_wrap_socket_impl
    return ssl_context.wrap_socket(sock, server_hostname=server_hostname)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "ssl.py", line 455, in wrap_socket
  File "ssl.py", line 1041, in _create
  File "ssl.py", line 1319, in do_handshake
urllib3.exceptions.ProtocolError: ('Connection aborted.', ConnectionResetError(10054, '远程主机强迫关闭了一个现有的连接。', None, 10054, None))

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\huggingface_hub\_snapshot_download.py", line 165, in snapshot_download
    repo_info = api.repo_info(repo_id=repo_id, repo_type=repo_type, revision=revision)
                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\huggingface_hub\utils\_validators.py", line 114, in _inner_fn
    return fn(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\huggingface_hub\hf_api.py", line 2864, in repo_info
    return method(
           ^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\huggingface_hub\utils\_validators.py", line 114, in _inner_fn
    return fn(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\huggingface_hub\hf_api.py", line 2648, in model_info
    r = get_session().get(path, headers=headers, timeout=timeout, params=params)
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\requests\sessions.py", line 602, in get
    return self.request("GET", url, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\requests\sessions.py", line 589, in request
    resp = self.send(prep, **send_kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\requests\sessions.py", line 703, in send
    r = adapter.send(request, **kwargs)
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\huggingface_hub\utils\_http.py", line 95, in send
    return super().send(request, *args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\requests\adapters.py", line 659, in send
    raise ConnectionError(err, request=request)
requests.exceptions.ConnectionError: (ProtocolError('Connection aborted.', ConnectionResetError(10054, '远程主机强迫关闭了一个现有的连接。', None, 10054, None)), '(Request ID: fdf6b831-3e6c-413e-8335-3c974bc5d7aa)')

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 496, in execute
    output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
                                                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 315, in get_output_data
    return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 289, in _async_map_node_over_list
    await process_inputs(input_dict, i)
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 277, in process_inputs
    result = f(**inputs)
             ^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-GIMM-VFI\nodes.py", line 67, in loadmodel
    snapshot_download(
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\huggingface_hub\utils\_validators.py", line 114, in _inner_fn
    return fn(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\huggingface_hub\_snapshot_download.py", line 248, in snapshot_download
    raise LocalEntryNotFoundError(
huggingface_hub.errors.LocalEntryNotFoundError: An error happened while trying to locate the files on the Hub and we cannot find the appropriate snapshot folder for the specified revision on the local disk. Please check your internet connection and try again.

2025-10-11T13:43:12.047253 - Prompt executed in 5.08 seconds
2025-10-11T13:43:41.871501 - got prompt
2025-10-11T13:43:41.957935 - Downloading GMMI-VFI model to: D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\models\interpolation\gimm-vfi\gimmvfi_r_arb_lpips_fp32.safetensors
2025-10-11T13:43:42.699216 - 
Fetching 1 files:   0%|                                                                          | 0/1 [00:00<?, ?it/s]2025-10-11T13:43:42.700216 - D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\huggingface_hub\file_download.py:982: UserWarning: `local_dir_use_symlinks` parameter is deprecated and will be ignored. The process to download files to a local folder has been updated and do not rely on symlinks anymore. You only need to pass a destination folder as`local_dir`.
For more details, check out https://huggingface.co/docs/huggingface_hub/main/en/guides/download#download-files-to-local-folder.
  warnings.warn(

2025-10-11T13:43:43.743376 - Xet Storage is enabled for this repo, but the 'hf_xet' package is not installed. Falling back to regular HTTP download. For better performance, install the package with: `pip install huggingface_hub[hf_xet]` or `pip install hf_xet`
2025-10-11T13:43:43.743376 - Xet Storage is enabled for this repo, but the 'hf_xet' package is not installed. Falling back to regular HTTP download. For better performance, install the package with: `pip install huggingface_hub[hf_xet]` or `pip install hf_xet`
2025-10-11T13:43:46.889730 - 
Fetching 1 files: 100%|██████████████████████████████████████████████████████████████████| 1/1 [00:04<00:00,  4.19s/it]2025-10-11T13:43:46.889730 - 
Fetching 1 files: 100%|██████████████████████████████████████████████████████████████████| 1/1 [00:04<00:00,  4.19s/it]2025-10-11T13:43:46.890736 - 
2025-10-11T13:43:46.891736 - Downloading RAFT model to: D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\models\interpolation\gimm-vfi\raft-things_fp32.safetensors
2025-10-11T13:43:47.185245 - 
Fetching 1 files:   0%|                                                                          | 0/1 [00:00<?, ?it/s]2025-10-11T13:43:48.153886 - Xet Storage is enabled for this repo, but the 'hf_xet' package is not installed. Falling back to regular HTTP download. For better performance, install the package with: `pip install huggingface_hub[hf_xet]` or `pip install hf_xet`
2025-10-11T13:43:48.155400 - Xet Storage is enabled for this repo, but the 'hf_xet' package is not installed. Falling back to regular HTTP download. For better performance, install the package with: `pip install huggingface_hub[hf_xet]` or `pip install hf_xet`
2025-10-11T13:43:52.997082 - 
Fetching 1 files: 100%|██████████████████████████████████████████████████████████████████| 1/1 [00:05<00:00,  5.81s/it]2025-10-11T13:43:53.003086 - 
Fetching 1 files: 100%|██████████████████████████████████████████████████████████████████| 1/1 [00:05<00:00,  5.82s/it]2025-10-11T13:43:53.003086 - 
2025-10-11T13:44:07.082629 - 
T5Encoder:   4%|███                                                                     | 1/24 [00:00<00:02,  9.30it/s]2025-10-11T13:44:07.143857 - 
T5Encoder: 100%|██████████████████████████████████████████████████████████████████████| 24/24 [00:00<00:00, 142.18it/s]2025-10-11T13:44:07.143857 - 
2025-10-11T13:44:07.144857 - prompt token count:2025-10-11T13:44:07.144857 -  2025-10-11T13:44:07.152857 - tensor([10], device='cuda:0')2025-10-11T13:44:07.152857 - 
2025-10-11T13:44:07.154857 - 
T5Encoder:   0%|                                                                                | 0/24 [00:00<?, ?it/s]2025-10-11T13:44:07.215326 - 
T5Encoder: 100%|██████████████████████████████████████████████████████████████████████| 24/24 [00:00<00:00, 396.90it/s]2025-10-11T13:44:07.215326 - 
2025-10-11T13:44:07.216326 - prompt token count:2025-10-11T13:44:07.216326 -  2025-10-11T13:44:07.224723 - tensor([126], device='cuda:0')2025-10-11T13:44:07.224723 - 
2025-10-11T13:44:11.101534 - Requested to load CLIPVisionModelProjection
2025-10-11T13:44:11.113230 - loaded completely 19659.561516571044 1208.09814453125 True
2025-10-11T13:44:11.232616 - Clip embeds shape: torch.Size([1, 257, 1280]), dtype: torch.float32
2025-10-11T13:44:11.232616 - Combined clip embeds shape: torch.Size([1, 257, 1280])
2025-10-11T13:44:24.904689 - !!! Exception during processing !!! Can't import SageAttention: No module named 'sageattention'
2025-10-11T13:44:24.907689 - Traceback (most recent call last):
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-WanVideoWrapper\nodes_model_loading.py", line 1037, in loadmodel
    from sageattention import sageattn
ModuleNotFoundError: No module named 'sageattention'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 496, in execute
    output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
                                                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 315, in get_output_data
    return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 289, in _async_map_node_over_list
    await process_inputs(input_dict, i)
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 277, in process_inputs
    result = f(**inputs)
             ^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-WanVideoWrapper\nodes_model_loading.py", line 1039, in loadmodel
    raise ValueError(f"Can't import SageAttention: {str(e)}")
ValueError: Can't import SageAttention: No module named 'sageattention'

2025-10-11T13:44:24.909689 - Prompt executed in 43.01 seconds
2025-10-11T13:44:43.531249 - got prompt
2025-10-11T13:44:43.630624 - !!! Exception during processing !!! Can't import SageAttention: No module named 'sageattention'
2025-10-11T13:44:43.631625 - Traceback (most recent call last):
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-WanVideoWrapper\nodes_model_loading.py", line 1037, in loadmodel
    from sageattention import sageattn
ModuleNotFoundError: No module named 'sageattention'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 496, in execute
    output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
                                                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 315, in get_output_data
    return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 289, in _async_map_node_over_list
    await process_inputs(input_dict, i)
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 277, in process_inputs
    result = f(**inputs)
             ^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-WanVideoWrapper\nodes_model_loading.py", line 1039, in loadmodel
    raise ValueError(f"Can't import SageAttention: {str(e)}")
ValueError: Can't import SageAttention: No module named 'sageattention'

2025-10-11T13:44:43.633926 - Prompt executed in 0.08 seconds
2025-10-11T13:49:24.350781 - got prompt
2025-10-11T13:49:24.508998 - Exception in callback _ProactorBasePipeTransport._call_connection_lost(None)
handle: <Handle _ProactorBasePipeTransport._call_connection_lost(None)>
Traceback (most recent call last):
  File "asyncio\events.py", line 88, in _run
  File "asyncio\proactor_events.py", line 165, in _call_connection_lost
ConnectionResetError: [WinError 10054] 远程主机强迫关闭了一个现有的连接。
2025-10-11T13:49:24.640291 - CUDA Compute Capability: 8.9
2025-10-11T13:49:24.640291 - !!! Exception during processing !!! The model is a scaled fp8 model, please set quantization to '_scaled'
2025-10-11T13:49:24.641292 - Traceback (most recent call last):
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 496, in execute
    output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
                                                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 315, in get_output_data
    return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 289, in _async_map_node_over_list
    await process_inputs(input_dict, i)
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 277, in process_inputs
    result = f(**inputs)
             ^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-WanVideoWrapper\nodes_model_loading.py", line 1110, in loadmodel
    raise ValueError("The model is a scaled fp8 model, please set quantization to '_scaled'")
ValueError: The model is a scaled fp8 model, please set quantization to '_scaled'

2025-10-11T13:49:24.642291 - Prompt executed in 0.26 seconds
2025-10-11T13:49:41.381292 - got prompt
2025-10-11T13:49:41.549489 - CUDA Compute Capability: 8.9
2025-10-11T13:49:41.549489 - !!! Exception during processing !!! The model is a scaled fp8 model, please set quantization to '_scaled'
2025-10-11T13:49:41.550489 - Traceback (most recent call last):
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 496, in execute
    output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
                                                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 315, in get_output_data
    return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 289, in _async_map_node_over_list
    await process_inputs(input_dict, i)
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 277, in process_inputs
    result = f(**inputs)
             ^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-WanVideoWrapper\nodes_model_loading.py", line 1110, in loadmodel
    raise ValueError("The model is a scaled fp8 model, please set quantization to '_scaled'")
ValueError: The model is a scaled fp8 model, please set quantization to '_scaled'

2025-10-11T13:49:41.552490 - Prompt executed in 0.14 seconds
2025-10-11T13:49:46.459624 - got prompt
2025-10-11T13:49:46.629463 - CUDA Compute Capability: 8.9
2025-10-11T13:49:46.629463 - !!! Exception during processing !!! The model is a scaled fp8 model, please set quantization to '_scaled'
2025-10-11T13:49:46.630463 - Traceback (most recent call last):
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 496, in execute
    output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
                                                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 315, in get_output_data
    return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 289, in _async_map_node_over_list
    await process_inputs(input_dict, i)
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 277, in process_inputs
    result = f(**inputs)
             ^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-WanVideoWrapper\nodes_model_loading.py", line 1110, in loadmodel
    raise ValueError("The model is a scaled fp8 model, please set quantization to '_scaled'")
ValueError: The model is a scaled fp8 model, please set quantization to '_scaled'

2025-10-11T13:49:46.633463 - Prompt executed in 0.15 seconds
2025-10-11T13:50:02.851230 - got prompt
2025-10-11T13:50:03.014792 - CUDA Compute Capability: 8.9
2025-10-11T13:50:03.014792 - Detected model in_channels: 36
2025-10-11T13:50:03.014792 - Model cross attention type: i2v, num_heads: 40, num_layers: 40
2025-10-11T13:50:03.016301 - Model variant detected: 14B
2025-10-11T13:50:03.654764 - model_type FLOW
2025-10-11T13:50:04.239709 - Loading LoRA: WanAnimate_relight_lora_fp16 with strength: 1.0
2025-10-11T13:50:04.442480 - Loading LoRA: lightx2v_I2V_14B_480p_cfg_step_distill_rank64_bf16 with strength: 1.2
2025-10-11T13:50:04.653252 - Loading LoRA: Wan22_PusaV1_lora_LOW_resized_dynamic_avg_rank_98_bf16 with strength: 1.0
2025-10-11T13:50:04.863773 - Using accelerate to load and assign model weights to device...
2025-10-11T13:50:05.175622 - 
Loading transformer parameters to cuda:0:  96%|████████████████████████████████▍ | 1363/1427 [00:00<00:00, 5494.28it/s]2025-10-11T13:50:05.347804 - 
Loading transformer parameters to cuda:0: 100%|██████████████████████████████████| 1427/1427 [00:00<00:00, 2957.35it/s]2025-10-11T13:50:05.347804 - 
2025-10-11T13:50:05.347804 - Using 1261 LoRA weight patches for WanVideo model
2025-10-11T13:50:05.409306 - ------- Scheduler info -------
2025-10-11T13:50:05.410307 - Total timesteps: tensor([999, 961, 909, 833, 714, 499], device='cuda:0')
2025-10-11T13:50:05.411499 - Using timesteps: tensor([999, 961, 909, 833, 714, 499], device='cuda:0')
2025-10-11T13:50:05.411499 - Using sigmas: tensor([1.0000, 0.9615, 0.9090, 0.8333, 0.7142, 0.4999, 0.0000])
2025-10-11T13:50:05.412499 - ------------------------------
2025-10-11T13:50:05.412499 - sigmas: tensor([1.0000, 0.9615, 0.9090, 0.8333, 0.7142, 0.4999, 0.0000])
2025-10-11T13:50:05.851084 - Input sequence length: 52650
2025-10-11T13:50:05.855144 - Sampling 101 frames at 720x720 with 6 steps
2025-10-11T13:50:06.296374 - 
  0%|                                                                                            | 0/6 [00:00<?, ?it/s]2025-10-11T13:50:09.918289 - Error during model prediction: Cannot find a working triton installation. Either the package is not installed or it is too old. More information on installing Triton can be found at: https://github.com/triton-lang/triton

Set TORCHDYNAMO_VERBOSE=1 for the internal stack trace (please do this especially if you're reporting a bug to PyTorch). For even more developer context, set TORCH_LOGS="+dynamo"

2025-10-11T13:50:10.451040 - 
  0%|                                                                                            | 0/6 [00:04<?, ?it/s]2025-10-11T13:50:10.451040 - 
2025-10-11T13:50:10.452040 - Error during sampling: Cannot find a working triton installation. Either the package is not installed or it is too old. More information on installing Triton can be found at: https://github.com/triton-lang/triton

Set TORCHDYNAMO_VERBOSE=1 for the internal stack trace (please do this especially if you're reporting a bug to PyTorch). For even more developer context, set TORCH_LOGS="+dynamo"

2025-10-11T13:50:10.992877 - !!! Exception during processing !!! Cannot find a working triton installation. Either the package is not installed or it is too old. More information on installing Triton can be found at: https://github.com/triton-lang/triton

Set TORCHDYNAMO_VERBOSE=1 for the internal stack trace (please do this especially if you're reporting a bug to PyTorch). For even more developer context, set TORCH_LOGS="+dynamo"

2025-10-11T13:50:10.997876 - Traceback (most recent call last):
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 496, in execute
    output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
                                                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 315, in get_output_data
    return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 289, in _async_map_node_over_list
    await process_inputs(input_dict, i)
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 277, in process_inputs
    result = f(**inputs)
             ^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-WanVideoWrapper\nodes_sampler.py", line 2906, in process
    raise e
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-WanVideoWrapper\nodes_sampler.py", line 2796, in process
    noise_pred, self.cache_state = predict_with_cfg(
                                   ^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-WanVideoWrapper\nodes_sampler.py", line 1445, in predict_with_cfg
    raise e
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-WanVideoWrapper\nodes_sampler.py", line 1318, in predict_with_cfg
    noise_pred_cond, cache_state_cond = transformer(
                                        ^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1773, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1784, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-WanVideoWrapper\wanvideo\modules\model.py", line 2708, in forward
    x, x_ip, lynx_ref_feature = block(x, x_ip=x_ip, lynx_ref_feature=lynx_ref_feature, **kwargs) #run block
                                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_dynamo\eval_frame.py", line 375, in __call__
    return super().__call__(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1773, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1784, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_dynamo\eval_frame.py", line 749, in compile_wrapper
    raise e.remove_dynamo_frames() from None  # see TORCHDYNAMO_VERBOSE=1
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\scheduler.py", line 4042, in create_backend
    raise TritonMissing(inspect.currentframe())
torch._inductor.exc.TritonMissing: Cannot find a working triton installation. Either the package is not installed or it is too old. More information on installing Triton can be found at: https://github.com/triton-lang/triton

Set TORCHDYNAMO_VERBOSE=1 for the internal stack trace (please do this especially if you're reporting a bug to PyTorch). For even more developer context, set TORCH_LOGS="+dynamo"


2025-10-11T13:50:11.005264 - Prompt executed in 8.13 seconds
2025-10-11T13:52:02.280180 - D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-mixlab-nodes\webApp\lib/photoswipe-lightbox.esm.min.js2025-10-11T13:52:02.281034 - 
2025-10-11T13:52:02.288348 - D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-mixlab-nodes\webApp\lib/photoswipe.min.css2025-10-11T13:52:02.289348 - 
2025-10-11T13:52:02.299349 - D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-mixlab-nodes\webApp\lib/pickr.min.js2025-10-11T13:52:02.299349 - 
2025-10-11T13:52:02.357711 - D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-mixlab-nodes\webApp\lib/classic.min.css2025-10-11T13:52:02.357711 - 
2025-10-11T13:52:02.360354 - D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-mixlab-nodes\webApp\lib/model-viewer.min.js2025-10-11T13:52:02.360354 - 
2025-10-11T13:52:02.388360 - D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-mixlab-nodes\webApp\lib/juxtapose.css2025-10-11T13:52:02.388360 - 
2025-10-11T13:52:02.390360 - D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-mixlab-nodes\webApp\lib/juxtapose.min.js2025-10-11T13:52:02.391359 - 
2025-10-11T13:52:05.059499 - Error reading JSON from D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\my_workflows\tushengtu_comfy 4.json: Expecting value: line 1 column 1 (char 0)2025-10-11T13:52:05.059499 - 
2025-10-11T13:52:06.685889 - Error scan workflow D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\my_workflows\tushengtu_comfy 4.json: Expecting value: line 1 column 1 (char 0), Traceback (most recent call last):
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-workspace-manager\service\scan_my_workflows_folder.py", line 35, in folder_handle
    file_handle(item, fileList, item_path, metaInfoOnly)
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-workspace-manager\service\scan_my_workflows_folder.py", line 55, in file_handle
    json_data = json.load(f)
                ^^^^^^^^^^^^
  File "json\__init__.py", line 293, in load
  File "json\__init__.py", line 346, in loads
  File "json\decoder.py", line 338, in decode
  File "json\decoder.py", line 356, in raw_decode
json.decoder.JSONDecodeError: Expecting value: line 1 column 1 (char 0)

2025-10-11T13:52:06.921221 - Error scan workflow D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\my_workflows\tushengtu_comfy 4.json: Expecting value: line 1 column 1 (char 0), Traceback (most recent call last):
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-workspace-manager\service\scan_my_workflows_folder.py", line 35, in folder_handle
    file_handle(item, fileList, item_path, metaInfoOnly)
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-workspace-manager\service\scan_my_workflows_folder.py", line 55, in file_handle
    json_data = json.load(f)
                ^^^^^^^^^^^^
  File "json\__init__.py", line 293, in load
  File "json\__init__.py", line 346, in loads
  File "json\decoder.py", line 338, in decode
  File "json\decoder.py", line 356, in raw_decode
json.decoder.JSONDecodeError: Expecting value: line 1 column 1 (char 0)

2025-10-11T13:52:13.871764 - Error scan workflow D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\my_workflows\tushengtu_comfy 4.json: Expecting value: line 1 column 1 (char 0), Traceback (most recent call last):
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-workspace-manager\service\scan_my_workflows_folder.py", line 35, in folder_handle
    file_handle(item, fileList, item_path, metaInfoOnly)
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-workspace-manager\service\scan_my_workflows_folder.py", line 55, in file_handle
    json_data = json.load(f)
                ^^^^^^^^^^^^
  File "json\__init__.py", line 293, in load
  File "json\__init__.py", line 346, in loads
  File "json\decoder.py", line 338, in decode
  File "json\decoder.py", line 356, in raw_decode
json.decoder.JSONDecodeError: Expecting value: line 1 column 1 (char 0)

2025-10-11T13:52:14.130498 - Error scan workflow D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\my_workflows\tushengtu_comfy 4.json: Expecting value: line 1 column 1 (char 0), Traceback (most recent call last):
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-workspace-manager\service\scan_my_workflows_folder.py", line 35, in folder_handle
    file_handle(item, fileList, item_path, metaInfoOnly)
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-workspace-manager\service\scan_my_workflows_folder.py", line 55, in file_handle
    json_data = json.load(f)
                ^^^^^^^^^^^^
  File "json\__init__.py", line 293, in load
  File "json\__init__.py", line 346, in loads
  File "json\decoder.py", line 338, in decode
  File "json\decoder.py", line 356, in raw_decode
json.decoder.JSONDecodeError: Expecting value: line 1 column 1 (char 0)

2025-10-11T13:52:20.778175 - got prompt
2025-10-11T13:52:54.396257 -     await asyncio.gather(*tasks)2025-10-11T13:52:54.396257 - 
2025-10-11T13:52:54.396257 -     async for message in websocket:2025-10-11T13:52:54.396257 - 
2025-10-11T13:52:54.396257 -     yield await self.recv()2025-10-11T13:52:54.396257 - 
2025-10-11T13:52:54.396257 -     raise self.protocol.close_exc from self.recv_exc2025-10-11T13:52:54.396257 - 
2025-10-11T13:52:54.397257 - Traceback (most recent call last):
2025-10-11T13:52:54.397257 - 2025-10-11T13:52:54.397257 -   File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_Bxb\wss.py", line 171, in websocket_connect
    await asyncio.gather(*tasks)
2025-10-11T13:52:54.397257 - 2025-10-11T13:52:54.397257 -   File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_Bxb\wss.py", line 498, in receive_messages
    async for message in websocket:
2025-10-11T13:52:54.397257 - 2025-10-11T13:52:54.397257 -   File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\websockets\asyncio\connection.py", line 242, in __aiter__
    yield await self.recv()
          ^^^^^^^^^^^^^^^^^
2025-10-11T13:52:54.397257 - 2025-10-11T13:52:54.397257 -   File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\websockets\asyncio\connection.py", line 322, in recv
    raise self.protocol.close_exc from self.recv_exc
2025-10-11T13:52:54.397257 - 2025-10-11T13:52:54.397257 - websockets.exceptions.ConnectionClosedError: no close frame received or sent
2025-10-11T13:52:54.398257 - 2025-10-11T13:52:55.564361 - 1 连接成功2025-10-11T13:52:55.566360 - 
2025-10-11T13:53:05.666749 - 3 连接成功2025-10-11T13:53:05.666749 - 
2025-10-11T13:56:30.693833 - Prompt executed in 249.91 seconds
2025-10-11T14:00:10.718765 - got prompt
2025-10-11T14:00:17.103243 - D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\python_embeded\Lib\site-packages\onnxruntime\capi\onnxruntime_inference_collection.py:123: UserWarning: Specified provider 'CUDAExecutionProvider' is not in available provider names.Available providers: 'AzureExecutionProvider, CPUExecutionProvider'
  warnings.warn(

2025-10-11T14:01:17.952128 - 
Detecting bboxes: 100%|████████████████████████████████████████████████████████████████| 97/97 [00:58<00:00,  1.65it/s]2025-10-11T14:01:17.955094 - 
Detecting bboxes: 100%|████████████████████████████████████████████████████████████████| 97/97 [00:58<00:00,  1.65it/s]2025-10-11T14:01:17.959355 - 
2025-10-11T14:02:38.814762 - 
Extracting keypoints: 100%|████████████████████████████████████████████████████████████| 97/97 [01:20<00:00,  1.22it/s]2025-10-11T14:02:38.817259 - 
Extracting keypoints: 100%|████████████████████████████████████████████████████████████| 97/97 [01:20<00:00,  1.20it/s]2025-10-11T14:02:38.823760 - 
2025-10-11T14:02:39.864506 - 
Drawing pose images:  82%|█████████████████████████████████████████████████▍          | 80/97 [00:00<00:00, 213.87it/s]2025-10-11T14:02:39.947432 - 
Drawing pose images: 100%|████████████████████████████████████████████████████████████| 97/97 [00:00<00:00, 218.22it/s]2025-10-11T14:02:39.948737 - 
2025-10-11T14:02:47.784253 - model_path: 2025-10-11T14:02:47.784253 -  2025-10-11T14:02:47.784253 - D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\models\sam2\sam2_hiera_base_plus.safetensors2025-10-11T14:02:47.784253 - 
2025-10-11T14:02:47.784253 - Using model config: D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-segment-anything-2\sam2_configs\sam2_hiera_b+.yaml2025-10-11T14:02:47.784253 - 
2025-10-11T14:02:48.729398 - Resizing to model input image size: 2025-10-11T14:02:48.729398 -  2025-10-11T14:02:48.729398 - 10242025-10-11T14:02:48.730398 - 
2025-10-11T14:02:50.671975 - combined labels: 2025-10-11T14:02:50.671975 -  2025-10-11T14:02:50.672975 - [1. 1. 1. 1. 1. 1. 1. 1.]2025-10-11T14:02:50.672975 - 
2025-10-11T14:02:50.672975 - combined labels shape: 2025-10-11T14:02:50.672975 -  2025-10-11T14:02:50.672975 - (8,)2025-10-11T14:02:50.674018 - 
2025-10-11T14:02:56.397217 - 
propagate in video: 100%|██████████████████████████████████████████████████████████████| 97/97 [00:05<00:00, 17.42it/s]2025-10-11T14:02:56.397217 - 
propagate in video: 100%|██████████████████████████████████████████████████████████████| 97/97 [00:05<00:00, 18.94it/s]2025-10-11T14:02:56.397217 - 
2025-10-11T14:03:01.822917 - 
BlockifyMask batch:  88%|█████████████████████████████████████████████████████▍       | 85/97 [00:00<00:00, 166.87it/s]2025-10-11T14:03:01.881264 - 
BlockifyMask batch: 100%|█████████████████████████████████████████████████████████████| 97/97 [00:00<00:00, 171.11it/s]2025-10-11T14:03:01.881264 - 
2025-10-11T14:03:03.872440 - 
DrawMaskOnImage batch:  99%|██████████████████████████████████████████████████████████▍| 96/97 [00:01<00:00, 54.29it/s]2025-10-11T14:03:03.887439 - 
DrawMaskOnImage batch: 100%|███████████████████████████████████████████████████████████| 97/97 [00:01<00:00, 54.50it/s]2025-10-11T14:03:03.888416 - 
2025-10-11T14:03:13.226305 - 
T5Encoder:  75%|████████████████████████████████████████████████████▌                 | 18/24 [00:00<00:00, 105.46it/s]2025-10-11T14:03:13.433956 - 
T5Encoder: 100%|███████████████████████████████████████████████████████████████████████| 24/24 [00:00<00:00, 63.27it/s]2025-10-11T14:03:13.434956 - 
2025-10-11T14:03:13.434956 - prompt token count:2025-10-11T14:03:13.434956 -  2025-10-11T14:03:13.469156 - tensor([10], device='cuda:0')2025-10-11T14:03:13.470158 - 
2025-10-11T14:03:13.471158 - 
T5Encoder:   0%|                                                                                | 0/24 [00:00<?, ?it/s]2025-10-11T14:03:13.531738 - 
T5Encoder: 100%|██████████████████████████████████████████████████████████████████████| 24/24 [00:00<00:00, 396.17it/s]2025-10-11T14:03:13.532738 - 
2025-10-11T14:03:13.532738 - prompt token count:2025-10-11T14:03:13.533738 -  2025-10-11T14:03:13.537738 - tensor([126], device='cuda:0')2025-10-11T14:03:13.537738 - 
2025-10-11T14:03:16.531228 - Requested to load CLIPVisionModelProjection
2025-10-11T14:03:16.545228 - loaded completely 19619.561516571044 1208.09814453125 True
2025-10-11T14:03:16.616293 - Clip embeds shape: torch.Size([1, 257, 1280]), dtype: torch.float32
2025-10-11T14:03:16.617292 - Combined clip embeds shape: torch.Size([1, 257, 1280])
2025-10-11T14:03:29.449860 - CUDA Compute Capability: 8.9
2025-10-11T14:03:29.449860 - Detected model in_channels: 36
2025-10-11T14:03:29.449860 - Model cross attention type: i2v, num_heads: 40, num_layers: 40
2025-10-11T14:03:29.450857 - Model variant detected: 14B
2025-10-11T14:03:29.992689 - model_type FLOW
2025-10-11T14:03:30.566266 - Loading LoRA: WanAnimate_relight_lora_fp16 with strength: 1.0
2025-10-11T14:03:30.632457 - Loading LoRA: lightx2v_I2V_14B_480p_cfg_step_distill_rank64_bf16 with strength: 1.2
2025-10-11T14:03:30.712543 - Loading LoRA: Wan22_PusaV1_lora_LOW_resized_dynamic_avg_rank_98_bf16 with strength: 1.0
2025-10-11T14:03:30.796955 - Using accelerate to load and assign model weights to device...
2025-10-11T14:03:31.014849 - 
Loading transformer parameters to cuda:0:  98%|█████████████████████████████████▏| 1392/1427 [00:00<00:00, 7567.94it/s]2025-10-11T14:03:31.087439 - 
Loading transformer parameters to cuda:0: 100%|██████████████████████████████████| 1427/1427 [00:00<00:00, 4923.18it/s]2025-10-11T14:03:31.087775 - 
2025-10-11T14:03:31.088916 - Using 1261 LoRA weight patches for WanVideo model
2025-10-11T14:03:31.160844 - ------- Scheduler info -------
2025-10-11T14:03:31.164847 - Total timesteps: tensor([1000.0000,  952.6630,  883.2125,  771.4307,  561.6498,   24.4141], device='cuda:0')
2025-10-11T14:03:31.167847 - Using timesteps: tensor([1000.0000,  952.6630,  883.2125,  771.4307,  561.6498,   24.4141], device='cuda:0')
2025-10-11T14:03:31.168847 - Using sigmas: tensor([1.0000, 0.9527, 0.8832, 0.7714, 0.5616, 0.0244, 0.0000], device='cuda:0')
2025-10-11T14:03:31.168847 - ------------------------------
2025-10-11T14:03:31.169847 - sigmas: tensor([1.0000, 0.9527, 0.8832, 0.7714, 0.5616, 0.0244, 0.0000], device='cuda:0')
2025-10-11T14:03:31.552279 - Input sequence length: 52650
2025-10-11T14:03:31.554280 - Sampling 101 frames at 720x720 with 6 steps
2025-10-11T14:03:31.847133 - 
  0%|                                                                                            | 0/6 [00:00<?, ?it/s]2025-10-11T14:03:33.157619 - Error during model prediction: Cannot find a working triton installation. Either the package is not installed or it is too old. More information on installing Triton can be found at: https://github.com/triton-lang/triton

Set TORCHDYNAMO_VERBOSE=1 for the internal stack trace (please do this especially if you're reporting a bug to PyTorch). For even more developer context, set TORCH_LOGS="+dynamo"

2025-10-11T14:03:33.451527 - 
  0%|                                                                                            | 0/6 [00:01<?, ?it/s]2025-10-11T14:03:33.451527 - 
2025-10-11T14:03:33.451527 - Error during sampling: Cannot find a working triton installation. Either the package is not installed or it is too old. More information on installing Triton can be found at: https://github.com/triton-lang/triton

Set TORCHDYNAMO_VERBOSE=1 for the internal stack trace (please do this especially if you're reporting a bug to PyTorch). For even more developer context, set TORCH_LOGS="+dynamo"

2025-10-11T14:03:33.936002 - !!! Exception during processing !!! Cannot find a working triton installation. Either the package is not installed or it is too old. More information on installing Triton can be found at: https://github.com/triton-lang/triton

Set TORCHDYNAMO_VERBOSE=1 for the internal stack trace (please do this especially if you're reporting a bug to PyTorch). For even more developer context, set TORCH_LOGS="+dynamo"

2025-10-11T14:03:33.941000 - Traceback (most recent call last):
  File "D:\ComfyUI_windows_portable_nvidia_cu128\ComfyUI_windows_portable\ComfyUI\execution.py", line 496, in execute
    output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
                                                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

[RuneGjerde]

Did you try install from a premade wheel ? its easier.
https://github.com/woct0rdho/triton-windows/releases/tag/v3.2.0-windows.post10

Download the one that matches your Python version... Cp312 = Python 3.12 etc

Activate your comfyui environment and write :pip installand then drop the file into windows terminal. .

See if that fixes it for you

After you have installed triton, you probably also want to install sage attention : pip install sage-attention

Edit: found a guide here that probably is easier to follow, since you are on portable you need the full python path before pip install https://www.patreon.com/posts/easy-guide-sage-124253103

[15036304416q-create]

Thank you very much, the problem is solved

[Jokcy]

@RuneGjerde Can you help with linux? I just followed comfyui manually install guide to setup comfyui on linux and start trying the mocha workflow, I got this error and I tried pip install sage-attention but it seems no sage-attention package there but sageattention exist. I tried to install sageattention but get this error:

Traceback (most recent call last):
        File "/home/oem/miniconda3/envs/comfyenv/lib/python3.12/site-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py", line 389, in <module>
          main()
        File "/home/oem/miniconda3/envs/comfyenv/lib/python3.12/site-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py", line 373, in main
          json_out["return_val"] = hook(**hook_input["kwargs"])
                                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
        File "/home/oem/miniconda3/envs/comfyenv/lib/python3.12/site-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py", line 143, in get_requires_for_build_wheel
          return hook(config_settings)
                 ^^^^^^^^^^^^^^^^^^^^^
        File "/tmp/pip-build-env-uh7gnzg2/overlay/lib/python3.12/site-packages/setuptools/build_meta.py", line 332, in get_requires_for_build_wheel
          return self._get_build_requires(config_settings, requirements=[])
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
        File "/tmp/pip-build-env-uh7gnzg2/overlay/lib/python3.12/site-packages/setuptools/build_meta.py", line 302, in _get_build_requires
          self.run_setup()
        File "/tmp/pip-build-env-uh7gnzg2/overlay/lib/python3.12/site-packages/setuptools/build_meta.py", line 318, in run_setup
          exec(code, locals())
        File "<string>", line 36, in <module>
      ModuleNotFoundError: No module named 'torch'

But if I try pip show torch it got:

Name: torch
Version: 2.6.0.dev20241112+cu121
Summary: Tensors and Dynamic neural networks in Python with strong GPU acceleration
Home-page: https://pytorch.org/
Author: PyTorch Team
Author-email: packages@pytorch.org
License: BSD-3-Clause
Location: /home/oem/miniconda3/envs/comfyenv/lib/python3.12/site-packages
Requires: filelock, fsspec, jinja2, networkx, nvidia-cublas-cu12, nvidia-cuda-cupti-cu12, nvidia-cuda-nvrtc-cu12, nvidia-cuda-runtime-cu12, nvidia-cudnn-cu12, nvidia-cufft-cu12, nvidia-curand-cu12, nvidia-cusolver-cu12, nvidia-cusparse-cu12, nvidia-cusparselt-cu12, nvidia-nccl-cu12, nvidia-nvtx-cu12, pytorch-triton, setuptools, sympy, typing-extensions
Required-by: accelerate, kornia, peft, spandrel, torchaudio, torchsde, torchvision

[RuneGjerde]

Activate the ComfyUI environment.

At the comfy folder in terminal write source venv/bin/activate (or the environment you set up, i see you might be using miniconda)

Then install triton first pip install triton followed by pip install sageattention (i think its written in one word if i remember right)

That should do it

---

Edit:
Should you have any compile errors, you could also try premade wheels.
I saw Kijai had some here https://huggingface.co/Kijai/PrecompiledWheels/tree/main
Triton & sage

[Jokcy]

Activate the ComfyUI environment.

At the comfy folder in terminal write source venv/bin/activate (or the environment you set up, i see you might be using miniconda)

Then install triton first pip install triton followed by pip install sageattention (i think its written in one word if i remember right)

That should do it

Edit: Should you have any compile errors, you could also try premade wheels. I saw Kijai had some here https://huggingface.co/Kijai/PrecompiledWheels/tree/main Triton & sage

THX, I have fix it by run pip install --no-build-isolation sageattention, seems it did not found cuda in global scope