cannot import name 'AttrsDescriptor' when using TorchCompileModel node

[edisonchan]

Expected Behavior

TorchCompileModel node run well.

Actual Behavior

popup a error message.

Steps to Reproduce

Add and connect TorchCompileModel node, and then click Queue button.

Debug Logs

OS: Windows 11
VS: 2022
cudatoolkit 12.8
ComfyUI version: 0.3.22
Python Version : 3.12.7 (tags/v3.12.7:0b05ead, Oct 1 2024, 03:06:41) [MSC v.1941 64 bit (AMD64)]
Pytorch Version: 2.6.0+cu128.nv
GPU: RTX 2070(SM7.5)

start comfyui with: ComfyUI\main.py --windows-standalone-build

# ComfyUI Error Report
## Error Details
- **Node ID:** 49
- **Node Type:** TorchCompileModel
- **Exception Type:** ImportError
- **Exception Message:** cannot import name 'AttrsDescriptor' from 'triton.compiler.compiler' (I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\triton\compiler\compiler.py)
## Stack Trace

  File "I:\ComfyUI_windows_portable\ComfyUI\execution.py", line 327, in execute
    output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
                                           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "I:\ComfyUI_windows_portable\ComfyUI\execution.py", line 202, in get_output_data
    return_values = _map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "I:\ComfyUI_windows_portable\ComfyUI\execution.py", line 174, in _map_node_over_list
    process_inputs(input_dict, i)

  File "I:\ComfyUI_windows_portable\ComfyUI\execution.py", line 163, in process_inputs
    results.append(getattr(obj, func)(**inputs))
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "I:\ComfyUI_windows_portable\ComfyUI\comfy_extras\nodes_torch_compile.py", line 17, in patch
    m.add_object_patch("diffusion_model", torch.compile(model=m.get_model_object("diffusion_model"), backend=backend))
                                          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\__init__.py", line 2520, in compile
    return torch._dynamo.optimize(
           ^^^^^^^^^^^^^^^^^^^^^^^

  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_dynamo\eval_frame.py", line 831, in optimize
    return _optimize(rebuild_ctx, *args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_dynamo\eval_frame.py", line 906, in _optimize
    backend.get_compiler_config()

  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\__init__.py", line 2304, in get_compiler_config
    from torch._inductor.compile_fx import get_patched_config_dict

  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\compile_fx.py", line 66, in <module>
    from torch._inductor.debug import save_args_for_compile_fx_inner

  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\debug.py", line 26, in <module>
    from . import config, ir  # noqa: F811, this is needed
    ^^^^^^^^^^^^^^^^^^^^^^^^

  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\ir.py", line 82, in <module>
    from .runtime.hints import ReductionHint

  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\runtime\hints.py", line 64, in <module>
    from triton.compiler.compiler import AttrsDescriptor


## System Information
- **ComfyUI Version:** 0.3.22
- **Arguments:** ComfyUI\main.py --windows-standalone-build --use-sage-attention
- **OS:** nt
- **Python Version:** 3.12.7 (tags/v3.12.7:0b05ead, Oct  1 2024, 03:06:41) [MSC v.1941 64 bit (AMD64)]
- **Embedded Python:** true
- **PyTorch Version:** 2.6.0+cu128.nv
## Devices

- **Name:** cuda:0 NVIDIA GeForce RTX 2070 : cudaMallocAsync
  - **Type:** cuda
  - **VRAM Total:** 8589475840
  - **VRAM Free:** 1622599160
  - **Torch VRAM Total:** 5100273664
  - **Torch VRAM Free:** 82142712

## Logs

2025-03-07T00:01:03.388004 - [START] Security scan2025-03-07T00:01:03.388004 - 
2025-03-07T00:01:04.015923 - [DONE] Security scan2025-03-07T00:01:04.015923 - 
2025-03-07T00:01:04.114046 - ## ComfyUI-Manager: installing dependencies done.2025-03-07T00:01:04.114046 - 
2025-03-07T00:01:04.114046 - ** ComfyUI startup time:2025-03-07T00:01:04.114046 -  2025-03-07T00:01:04.114046 - 2025-03-07 00:01:04.1142025-03-07T00:01:04.114046 - 
2025-03-07T00:01:04.114046 - ** Platform:2025-03-07T00:01:04.114046 -  2025-03-07T00:01:04.114046 - Windows2025-03-07T00:01:04.114046 - 
2025-03-07T00:01:04.114046 - ** Python version:2025-03-07T00:01:04.114046 -  2025-03-07T00:01:04.114046 - 3.12.7 (tags/v3.12.7:0b05ead, Oct  1 2024, 03:06:41) [MSC v.1941 64 bit (AMD64)]2025-03-07T00:01:04.114046 - 
2025-03-07T00:01:04.114046 - ** Python executable:2025-03-07T00:01:04.115319 -  2025-03-07T00:01:04.115319 - I:\ComfyUI_windows_portable\python_embeded\python.exe2025-03-07T00:01:04.115319 - 
2025-03-07T00:01:04.115319 - ** ComfyUI Path:2025-03-07T00:01:04.115319 -  2025-03-07T00:01:04.115319 - I:\ComfyUI_windows_portable\ComfyUI2025-03-07T00:01:04.115319 - 
2025-03-07T00:01:04.115319 - ** ComfyUI Base Folder Path:2025-03-07T00:01:04.115797 -  2025-03-07T00:01:04.115797 - I:\ComfyUI_windows_portable\ComfyUI2025-03-07T00:01:04.115797 - 
2025-03-07T00:01:04.115797 - ** User directory:2025-03-07T00:01:04.115797 -  2025-03-07T00:01:04.115797 - I:\ComfyUI_windows_portable\ComfyUI\user2025-03-07T00:01:04.115797 - 
2025-03-07T00:01:04.115797 - ** ComfyUI-Manager config path:2025-03-07T00:01:04.115797 -  2025-03-07T00:01:04.115797 - I:\ComfyUI_windows_portable\ComfyUI\user\default\ComfyUI-Manager\config.ini2025-03-07T00:01:04.115797 - 
2025-03-07T00:01:04.115797 - ** Log path:2025-03-07T00:01:04.115797 -  2025-03-07T00:01:04.115797 - I:\ComfyUI_windows_portable\ComfyUI\user\comfyui.log2025-03-07T00:01:04.115797 - 
2025-03-07T00:01:04.827007 - [ComfyUI-Manager] PyTorch is not installed
2025-03-07T00:01:04.828011 - 
Prestartup times for custom nodes:
2025-03-07T00:01:04.828011 -    0.0 seconds: I:\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-easy-use
2025-03-07T00:01:04.828011 -    2.0 seconds: I:\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-manager
2025-03-07T00:01:04.828011 - 
2025-03-07T00:01:06.131077 - Checkpoint files will always be loaded safely.
2025-03-07T00:01:06.271573 - Total VRAM 8192 MB, total RAM 32677 MB
2025-03-07T00:01:06.272578 - pytorch version: 2.6.0+cu128.nv
2025-03-07T00:01:06.272578 - Set vram state to: NORMAL_VRAM
2025-03-07T00:01:06.272578 - Device: cuda:0 NVIDIA GeForce RTX 2070 : cudaMallocAsync
2025-03-07T00:01:07.325365 - Using sage attention
2025-03-07T00:01:08.418496 - ComfyUI version: 0.3.22
2025-03-07T00:01:08.443716 - [Prompt Server] web root: I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\comfyui_frontend_package\static
2025-03-07T00:01:08.763369 - Traceback (most recent call last):
  File "I:\ComfyUI_windows_portable\ComfyUI\nodes.py", line 2147, in load_custom_node
    module_spec.loader.exec_module(module)
  File "<frozen importlib._bootstrap_external>", line 995, in exec_module
  File "<frozen importlib._bootstrap>", line 488, in _call_with_frames_removed
  File "I:\ComfyUI_windows_portable\ComfyUI\comfy_extras\nodes_audio.py", line 1, in <module>
    import torchaudio
ModuleNotFoundError: No module named 'torchaudio'

2025-03-07T00:01:08.763369 - Cannot import I:\ComfyUI_windows_portable\ComfyUI\comfy_extras\nodes_audio.py module for custom nodes: No module named 'torchaudio'
2025-03-07T00:01:09.408064 - �[34m[ComfyUI-Easy-Use] server: �[0mv1.2.8 �[92mLoaded�[0m2025-03-07T00:01:09.408064 - 
2025-03-07T00:01:09.409067 - �[34m[ComfyUI-Easy-Use] web root: �[0mI:\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-easy-use\web_version/v2 �[92mLoaded�[0m2025-03-07T00:01:09.409067 - 
2025-03-07T00:01:09.416921 - Total VRAM 8192 MB, total RAM 32677 MB
2025-03-07T00:01:09.416921 - pytorch version: 2.6.0+cu128.nv
2025-03-07T00:01:09.416921 - Set vram state to: NORMAL_VRAM
2025-03-07T00:01:09.417925 - Device: cuda:0 NVIDIA GeForce RTX 2070 : cudaMallocAsync
2025-03-07T00:01:09.444704 - ### Loading: ComfyUI-Manager (V3.30.2)
2025-03-07T00:01:09.444704 - [ComfyUI-Manager] network_mode: public
2025-03-07T00:01:09.524146 - ### ComfyUI Revision: 3211 [88951997] *DETACHED | Released on '2025-03-05'
2025-03-07T00:01:09.765768 - 
Import times for custom nodes:
2025-03-07T00:01:09.765768 -    0.0 seconds: I:\ComfyUI_windows_portable\ComfyUI\custom_nodes\websocket_image_save.py
2025-03-07T00:01:09.765768 -    0.0 seconds: I:\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-show-text
2025-03-07T00:01:09.765768 -    0.0 seconds: I:\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui_profilerx
2025-03-07T00:01:09.765768 -    0.0 seconds: I:\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-kjnodes
2025-03-07T00:01:09.765768 -    0.0 seconds: I:\ComfyUI_windows_portable\ComfyUI\custom_nodes\Kurdknight_comfycheck
2025-03-07T00:01:09.765768 -    0.3 seconds: I:\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-manager
2025-03-07T00:01:09.765768 -    0.6 seconds: I:\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-easy-use
2025-03-07T00:01:09.765768 - 
2025-03-07T00:01:09.765768 - WARNING: some comfy_extras/ nodes did not import correctly. This may be because they are missing some dependencies.

2025-03-07T00:01:09.765768 - IMPORT FAILED: nodes_audio.py
2025-03-07T00:01:09.765768 - 
This issue might be caused by new missing dependencies added the last time you updated ComfyUI.
2025-03-07T00:01:09.765768 - Please run the update script: update/update_comfyui.bat
2025-03-07T00:01:09.765768 - 
2025-03-07T00:01:09.774487 - Starting server

2025-03-07T00:01:09.774487 - To see the GUI go to: http://127.0.0.1:8188
2025-03-07T00:01:09.803059 - [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/model-list.json
2025-03-07T00:01:09.819509 - [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/github-stats.json
2025-03-07T00:01:09.823999 - [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/alter-list.json
2025-03-07T00:01:09.857827 - [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/extension-node-map.json
2025-03-07T00:01:09.884906 - [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/custom-node-list.json
2025-03-07T00:01:12.949582 - got prompt
2025-03-07T00:01:12.949582 - 2025-03-07 00:01:12,949 - ComfyUI-ProfilerX - DEBUG - Average reset_peak_memory_stats time: 2294.80 nanoseconds
2025-03-07T00:01:12.949582 - Average reset_peak_memory_stats time: 2294.80 nanoseconds
2025-03-07T00:01:12.981996 - Using pytorch attention in VAE
2025-03-07T00:01:12.984141 - Using pytorch attention in VAE
2025-03-07T00:01:13.109689 - VAE load device: cuda:0, offload device: cpu, dtype: torch.float32
2025-03-07T00:01:13.334939 - CLIP/text encoder model load device: cuda:0, offload device: cpu, current: cpu, dtype: torch.float16
2025-03-07T00:01:13.436478 - clip missing: ['text_projection.weight']
2025-03-07T00:01:13.701489 - FETCH ComfyRegistry Data: 5/362025-03-07T00:01:13.701489 - 
2025-03-07T00:01:15.860526 - Requested to load FluxClipModel_
2025-03-07T00:01:16.949727 - loaded completely 5714.8 4777.53759765625 True
2025-03-07T00:01:17.719593 - Token indices sequence length is longer than the specified maximum sequence length for this model (82 > 77). Running this sequence through the model will result in indexing errors
2025-03-07T00:01:18.517729 - FETCH ComfyRegistry Data: 10/362025-03-07T00:01:18.518884 - 
2025-03-07T00:01:18.617803 - model weight dtype torch.float8_e4m3fn, manual cast: torch.float16
2025-03-07T00:01:18.617803 - model_type FLUX
2025-03-07T00:01:22.902758 - FETCH ComfyRegistry Data: 15/362025-03-07T00:01:22.902758 - 
2025-03-07T00:01:23.021140 - Processing interrupted
2025-03-07T00:01:23.088141 - Prompt executed in 10.13 seconds
2025-03-07T00:01:24.071660 - 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。
2025-03-07T00:01:27.001913 - FETCH ComfyRegistry Data: 20/362025-03-07T00:01:27.001913 - 
2025-03-07T00:01:31.501022 - FETCH ComfyRegistry Data: 25/362025-03-07T00:01:31.501022 - 
2025-03-07T00:01:36.859595 - FETCH ComfyRegistry Data: 30/362025-03-07T00:01:36.859595 - 
2025-03-07T00:01:41.006346 - FETCH ComfyRegistry Data: 35/362025-03-07T00:01:41.006346 - 
2025-03-07T00:01:42.326900 - FETCH ComfyRegistry Data [DONE]2025-03-07T00:01:42.326900 - 
2025-03-07T00:01:42.366708 - [ComfyUI-Manager] default cache updated: https://api.comfy.org/nodes
2025-03-07T00:01:42.402915 - nightly_channel: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/remote
2025-03-07T00:01:42.402915 - FETCH DATA from: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/custom-node-list.json2025-03-07T00:01:42.402915 - 2025-03-07T00:01:42.491081 -  [DONE]2025-03-07T00:01:42.492084 - 
2025-03-07T00:01:42.519363 - [ComfyUI-Manager] All startup tasks have been completed.
2025-03-07T00:02:23.748528 - got prompt
2025-03-07T00:02:23.814790 - !!! Exception during processing !!! cannot import name 'AttrsDescriptor' from 'triton.compiler.compiler' (I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\triton\compiler\compiler.py)
2025-03-07T00:02:23.814790 - Traceback (most recent call last):
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\runtime\hints.py", line 43, in <module>
    from triton.backends.compiler import AttrsDescriptor
ImportError: cannot import name 'AttrsDescriptor' from 'triton.backends.compiler' (I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\triton\backends\compiler.py)

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "I:\ComfyUI_windows_portable\ComfyUI\execution.py", line 327, in execute
    output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
                                           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\ComfyUI\execution.py", line 202, in get_output_data
    return_values = _map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\ComfyUI\execution.py", line 174, in _map_node_over_list
    process_inputs(input_dict, i)
  File "I:\ComfyUI_windows_portable\ComfyUI\execution.py", line 163, in process_inputs
    results.append(getattr(obj, func)(**inputs))
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\ComfyUI\comfy_extras\nodes_torch_compile.py", line 17, in patch
    m.add_object_patch("diffusion_model", torch.compile(model=m.get_model_object("diffusion_model"), backend=backend))
                                          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\__init__.py", line 2520, in compile
    return torch._dynamo.optimize(
           ^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_dynamo\eval_frame.py", line 831, in optimize
    return _optimize(rebuild_ctx, *args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_dynamo\eval_frame.py", line 906, in _optimize
    backend.get_compiler_config()
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\__init__.py", line 2304, in get_compiler_config
    from torch._inductor.compile_fx import get_patched_config_dict
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\compile_fx.py", line 66, in <module>
    from torch._inductor.debug import save_args_for_compile_fx_inner
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\debug.py", line 26, in <module>
    from . import config, ir  # noqa: F811, this is needed
    ^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\ir.py", line 82, in <module>
    from .runtime.hints import ReductionHint
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\runtime\hints.py", line 64, in <module>
    from triton.compiler.compiler import AttrsDescriptor
ImportError: cannot import name 'AttrsDescriptor' from 'triton.compiler.compiler' (I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\triton\compiler\compiler.py)

2025-03-07T00:02:23.814790 - Prompt executed in 0.07 seconds
2025-03-07T00:02:52.341582 - got prompt
2025-03-07T00:02:52.351701 - !!! Exception during processing !!! cannot import name 'AttrsDescriptor' from 'triton.compiler.compiler' (I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\triton\compiler\compiler.py)
2025-03-07T00:02:52.358502 - Traceback (most recent call last):
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\runtime\hints.py", line 43, in <module>
    from triton.backends.compiler import AttrsDescriptor
ImportError: cannot import name 'AttrsDescriptor' from 'triton.backends.compiler' (I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\triton\backends\compiler.py)

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "I:\ComfyUI_windows_portable\ComfyUI\execution.py", line 327, in execute
    output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
                                           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\ComfyUI\execution.py", line 202, in get_output_data
    return_values = _map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\ComfyUI\execution.py", line 174, in _map_node_over_list
    process_inputs(input_dict, i)
  File "I:\ComfyUI_windows_portable\ComfyUI\execution.py", line 163, in process_inputs
    results.append(getattr(obj, func)(**inputs))
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\ComfyUI\comfy_extras\nodes_torch_compile.py", line 17, in patch
    m.add_object_patch("diffusion_model", torch.compile(model=m.get_model_object("diffusion_model"), backend=backend))
                                          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\__init__.py", line 2520, in compile
    return torch._dynamo.optimize(
           ^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_dynamo\eval_frame.py", line 831, in optimize
    return _optimize(rebuild_ctx, *args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_dynamo\eval_frame.py", line 906, in _optimize
    backend.get_compiler_config()
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\__init__.py", line 2304, in get_compiler_config
    from torch._inductor.compile_fx import get_patched_config_dict
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\compile_fx.py", line 66, in <module>
    from torch._inductor.debug import save_args_for_compile_fx_inner
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\debug.py", line 26, in <module>
    from . import config, ir  # noqa: F811, this is needed
    ^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\ir.py", line 82, in <module>
    from .runtime.hints import ReductionHint
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\runtime\hints.py", line 64, in <module>
    from triton.compiler.compiler import AttrsDescriptor
ImportError: cannot import name 'AttrsDescriptor' from 'triton.compiler.compiler' (I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\triton\compiler\compiler.py)

2025-03-07T00:02:52.361511 - Prompt executed in 0.01 seconds
2025-03-07T00:03:32.597859 - got prompt
2025-03-07T00:03:32.597859 - !!! Exception during processing !!! cannot import name 'AttrsDescriptor' from 'triton.compiler.compiler' (I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\triton\compiler\compiler.py)
2025-03-07T00:03:32.614998 - Traceback (most recent call last):
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\runtime\hints.py", line 43, in <module>
    from triton.backends.compiler import AttrsDescriptor
ImportError: cannot import name 'AttrsDescriptor' from 'triton.backends.compiler' (I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\triton\backends\compiler.py)

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "I:\ComfyUI_windows_portable\ComfyUI\execution.py", line 327, in execute
    output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
                                           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\ComfyUI\execution.py", line 202, in get_output_data
    return_values = _map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\ComfyUI\execution.py", line 174, in _map_node_over_list
    process_inputs(input_dict, i)
  File "I:\ComfyUI_windows_portable\ComfyUI\execution.py", line 163, in process_inputs
    results.append(getattr(obj, func)(**inputs))
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\ComfyUI\comfy_extras\nodes_torch_compile.py", line 17, in patch
    m.add_object_patch("diffusion_model", torch.compile(model=m.get_model_object("diffusion_model"), backend=backend))
                                          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\__init__.py", line 2520, in compile
    return torch._dynamo.optimize(
           ^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_dynamo\eval_frame.py", line 831, in optimize
    return _optimize(rebuild_ctx, *args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_dynamo\eval_frame.py", line 906, in _optimize
    backend.get_compiler_config()
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\__init__.py", line 2304, in get_compiler_config
    from torch._inductor.compile_fx import get_patched_config_dict
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\compile_fx.py", line 66, in <module>
    from torch._inductor.debug import save_args_for_compile_fx_inner
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\debug.py", line 26, in <module>
    from . import config, ir  # noqa: F811, this is needed
    ^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\ir.py", line 82, in <module>
    from .runtime.hints import ReductionHint
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\runtime\hints.py", line 64, in <module>
    from triton.compiler.compiler import AttrsDescriptor
ImportError: cannot import name 'AttrsDescriptor' from 'triton.compiler.compiler' (I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\triton\compiler\compiler.py)

2025-03-07T00:03:32.618010 - Prompt executed in 0.02 seconds
2025-03-07T00:03:46.600220 - got prompt
2025-03-07T00:03:46.600220 - !!! Exception during processing !!! cannot import name 'AttrsDescriptor' from 'triton.compiler.compiler' (I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\triton\compiler\compiler.py)
2025-03-07T00:03:46.614184 - Traceback (most recent call last):
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\runtime\hints.py", line 43, in <module>
    from triton.backends.compiler import AttrsDescriptor
ImportError: cannot import name 'AttrsDescriptor' from 'triton.backends.compiler' (I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\triton\backends\compiler.py)

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "I:\ComfyUI_windows_portable\ComfyUI\execution.py", line 327, in execute
    output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
                                           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\ComfyUI\execution.py", line 202, in get_output_data
    return_values = _map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\ComfyUI\execution.py", line 174, in _map_node_over_list
    process_inputs(input_dict, i)
  File "I:\ComfyUI_windows_portable\ComfyUI\execution.py", line 163, in process_inputs
    results.append(getattr(obj, func)(**inputs))
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\ComfyUI\comfy_extras\nodes_torch_compile.py", line 17, in patch
    m.add_object_patch("diffusion_model", torch.compile(model=m.get_model_object("diffusion_model"), backend=backend))
                                          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\__init__.py", line 2520, in compile
    return torch._dynamo.optimize(
           ^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_dynamo\eval_frame.py", line 831, in optimize
    return _optimize(rebuild_ctx, *args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_dynamo\eval_frame.py", line 906, in _optimize
    backend.get_compiler_config()
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\__init__.py", line 2304, in get_compiler_config
    from torch._inductor.compile_fx import get_patched_config_dict
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\compile_fx.py", line 66, in <module>
    from torch._inductor.debug import save_args_for_compile_fx_inner
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\debug.py", line 26, in <module>
    from . import config, ir  # noqa: F811, this is needed
    ^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\ir.py", line 82, in <module>
    from .runtime.hints import ReductionHint
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\runtime\hints.py", line 64, in <module>
    from triton.compiler.compiler import AttrsDescriptor
ImportError: cannot import name 'AttrsDescriptor' from 'triton.compiler.compiler' (I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\triton\compiler\compiler.py)

2025-03-07T00:03:46.614184 - Prompt executed in 0.01 seconds
2025-03-07T00:07:45.312043 - got prompt
2025-03-07T00:07:45.333193 - !!! Exception during processing !!! cannot import name 'AttrsDescriptor' from 'triton.compiler.compiler' (I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\triton\compiler\compiler.py)
2025-03-07T00:07:45.335202 - Traceback (most recent call last):
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\runtime\hints.py", line 43, in <module>
    from triton.backends.compiler import AttrsDescriptor
ImportError: cannot import name 'AttrsDescriptor' from 'triton.backends.compiler' (I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\triton\backends\compiler.py)

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "I:\ComfyUI_windows_portable\ComfyUI\execution.py", line 327, in execute
    output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
                                           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\ComfyUI\execution.py", line 202, in get_output_data
    return_values = _map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\ComfyUI\execution.py", line 174, in _map_node_over_list
    process_inputs(input_dict, i)
  File "I:\ComfyUI_windows_portable\ComfyUI\execution.py", line 163, in process_inputs
    results.append(getattr(obj, func)(**inputs))
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\ComfyUI\comfy_extras\nodes_torch_compile.py", line 17, in patch
    m.add_object_patch("diffusion_model", torch.compile(model=m.get_model_object("diffusion_model"), backend=backend))
                                          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\__init__.py", line 2520, in compile
    return torch._dynamo.optimize(
           ^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_dynamo\eval_frame.py", line 831, in optimize
    return _optimize(rebuild_ctx, *args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_dynamo\eval_frame.py", line 906, in _optimize
    backend.get_compiler_config()
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\__init__.py", line 2304, in get_compiler_config
    from torch._inductor.compile_fx import get_patched_config_dict
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\compile_fx.py", line 66, in <module>
    from torch._inductor.debug import save_args_for_compile_fx_inner
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\debug.py", line 26, in <module>
    from . import config, ir  # noqa: F811, this is needed
    ^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\ir.py", line 82, in <module>
    from .runtime.hints import ReductionHint
  File "I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\_inductor\runtime\hints.py", line 64, in <module>
    from triton.compiler.compiler import AttrsDescriptor
ImportError: cannot import name 'AttrsDescriptor' from 'triton.compiler.compiler' (I:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\triton\compiler\compiler.py)

2025-03-07T00:07:45.338226 - Prompt executed in 0.01 seconds


## Attached Workflow
Please make sure that workflow does not contain any sensitive information such as API keys or passwords.

{"last_node_id":49,"last_link_id":82,"nodes":[{"id":8,"type":"VAEDecode","pos":[1502.1177978515625,100.8399658203125],"size":[210,46],"flags":{},"order":10,"mode":0,"inputs":[{"name":"samples","localized_name":"samples","type":"LATENT","link":52},{"name":"vae","localized_name":"vae","type":"VAE","link":59}],"outputs":[{"name":"IMAGE","localized_name":"IMAGE","type":"IMAGE","links":[9],"slot_index":0}],"properties":{"cnr_id":"comfy-core","ver":"0.3.18","Node name for S&R":"VAEDecode"},"widgets_values":[]},{"id":35,"type":"FluxGuidance","pos":[880,-40],"size":[211.60000610351562,58],"flags":{},"order":8,"mode":0,"inputs":[{"name":"conditioning","localized_name":"conditioning","type":"CONDITIONING","link":56}],"outputs":[{"name":"CONDITIONING","localized_name":"CONDITIONING","type":"CONDITIONING","shape":3,"links":[57],"slot_index":0}],"properties":{"cnr_id":"comfy-core","ver":"0.3.18","Node name for S&R":"FluxGuidance"},"widgets_values":[3.5]},{"id":33,"type":"CLIPTextEncode","pos":[400,200],"size":[422.84503173828125,164.31304931640625],"flags":{"collapsed":true},"order":7,"mode":0,"inputs":[{"name":"clip","localized_name":"clip","type":"CLIP","link":65,"slot_index":0}],"outputs":[{"name":"CONDITIONING","localized_name":"CONDITIONING","type":"CONDITIONING","links":[55],"slot_index":0}],"title":"CLIP Text Encode (Negative Prompt)","properties":{"cnr_id":"comfy-core","ver":"0.3.18","Node name for S&R":"CLIPTextEncode"},"widgets_values":[""],"color":"#322","bgcolor":"#533"},{"id":6,"type":"CLIPTextEncode","pos":[400,-40],"size":[422.84503173828125,164.31304931640625],"flags":{"collapsed":false},"order":6,"mode":0,"inputs":[{"name":"clip","localized_name":"clip","type":"CLIP","link":64}],"outputs":[{"name":"CONDITIONING","localized_name":"CONDITIONING","type":"CONDITIONING","links":[56],"slot_index":0}],"title":"CLIP Text Encode (Positive Prompt)","properties":{"cnr_id":"comfy-core","ver":"0.3.18","Node name for S&R":"CLIPTextEncode"},"widgets_values":["cute anime girl with massive fluffy fennec ears and a big fluffy tail blonde messy long hair blue eyes wearing a maid outfit with a long black gold leaf pattern dress and a white apron mouth open placing a fancy black forest cake with candles on top of a dinner table of an old dark Victorian mansion lit by candlelight with a bright window to the foggy forest and very expensive stuff everywhere there are paintings on the walls"],"color":"#232","bgcolor":"#353"},{"id":34,"type":"Note","pos":[1120,420],"size":[282.8617858886719,164.08004760742188],"flags":{},"order":0,"mode":0,"inputs":[],"outputs":[],"properties":{"text":""},"widgets_values":["Note that Flux dev and schnell do not have any negative prompt so CFG should be set to 1.0. Setting CFG to 1.0 means the negative prompt is ignored."],"color":"#432","bgcolor":"#653"},{"id":37,"type":"UNETLoader","pos":[-100,-50],"size":[315,82],"flags":{},"order":1,"mode":0,"inputs":[],"outputs":[{"name":"MODEL","localized_name":"MODEL","type":"MODEL","links":[82],"slot_index":0}],"properties":{"cnr_id":"comfy-core","ver":"0.3.18","Node name for S&R":"UNETLoader"},"widgets_values":["FLUX.1\\flux1-dev-fp8.safetensors","fp8_e4m3fn_fast"]},{"id":41,"type":"DualCLIPLoader","pos":[-100,120],"size":[315,122],"flags":{},"order":2,"mode":0,"inputs":[],"outputs":[{"name":"CLIP","localized_name":"CLIP","type":"CLIP","links":[64,65]}],"properties":{"cnr_id":"comfy-core","ver":"0.3.18","Node name for S&R":"DualCLIPLoader"},"widgets_values":["FLUX.1\\clip_l.safetensors","FLUX.1\\t5xxl_fp8_e4m3fn.safetensors","flux","default"]},{"id":38,"type":"VAELoader","pos":[-100,310],"size":[315,58],"flags":{},"order":3,"mode":0,"inputs":[],"outputs":[{"name":"VAE","localized_name":"VAE","type":"VAE","links":[59,67]}],"properties":{"cnr_id":"comfy-core","ver":"0.3.18","Node name for S&R":"VAELoader"},"widgets_values":["FLUX.1\\ae.safetensors"]},{"id":42,"type":"VAEDecodeTiled","pos":[1466.1331787109375,-150.12379455566406],"size":[315,150],"flags":{},"order":11,"mode":4,"inputs":[{"name":"samples","localized_name":"samples","type":"LATENT","link":66},{"name":"vae","localized_name":"vae","type":"VAE","link":67}],"outputs":[{"name":"IMAGE","localized_name":"IMAGE","type":"IMAGE","links":null,"slot_index":0}],"properties":{"cnr_id":"comfy-core","ver":"0.3.22","Node name for S&R":"VAEDecodeTiled"},"widgets_values":[512,64,64,8]},{"id":9,"type":"SaveImage","pos":[1790,100],"size":[582.2000122070312,437.13336181640625],"flags":{},"order":12,"mode":0,"inputs":[{"name":"images","localized_name":"images","type":"IMAGE","link":9}],"outputs":[],"properties":{"cnr_id":"comfy-core","ver":"0.3.18","Node name for S&R":"SaveImage"},"widgets_values":["ComfyUI"]},{"id":31,"type":"KSampler","pos":[1120,100],"size":[315,262],"flags":{},"order":9,"mode":0,"inputs":[{"name":"model","localized_name":"model","type":"MODEL","link":81},{"name":"positive","localized_name":"positive","type":"CONDITIONING","link":57},{"name":"negative","localized_name":"negative","type":"CONDITIONING","link":55},{"name":"latent_image","localized_name":"latent_image","type":"LATENT","link":51}],"outputs":[{"name":"LATENT","localized_name":"LATENT","type":"LATENT","shape":3,"links":[52,66],"slot_index":0}],"properties":{"cnr_id":"comfy-core","ver":"0.3.18","Node name for S&R":"KSampler"},"widgets_values":[364900715038437,"randomize",20,1,"euler","simple",1]},{"id":27,"type":"EmptySD3LatentImage","pos":[490,260],"size":[315,106],"flags":{},"order":4,"mode":0,"inputs":[],"outputs":[{"name":"LATENT","localized_name":"LATENT","type":"LATENT","shape":3,"links":[51],"slot_index":0}],"properties":{"cnr_id":"comfy-core","ver":"0.3.18","Node name for S&R":"EmptySD3LatentImage"},"widgets_values":[1024,1024,1],"color":"#323","bgcolor":"#535"},{"id":49,"type":"TorchCompileModel","pos":[476.1605224609375,-290.28387451171875],"size":[315,58],"flags":{},"order":5,"mode":0,"inputs":[{"name":"model","localized_name":"model","type":"MODEL","link":82}],"outputs":[{"name":"MODEL","localized_name":"MODEL","type":"MODEL","links":[81],"slot_index":0}],"properties":{"cnr_id":"comfy-core","ver":"0.3.22","Node name for S&R":"TorchCompileModel"},"widgets_values":["inductor"]}],"links":[[9,8,0,9,0,"IMAGE"],[51,27,0,31,3,"LATENT"],[52,31,0,8,0,"LATENT"],[55,33,0,31,2,"CONDITIONING"],[56,6,0,35,0,"CONDITIONING"],[57,35,0,31,1,"CONDITIONING"],[59,38,0,8,1,"VAE"],[64,41,0,6,0,"CLIP"],[65,41,0,33,0,"CLIP"],[66,31,0,42,0,"LATENT"],[67,38,0,42,1,"VAE"],[81,49,0,31,0,"MODEL"],[82,37,0,49,0,"MODEL"]],"groups":[],"config":{},"extra":{"ds":{"scale":0.7513148009015777,"offset":[666.7248772718718,521.8779592773864]},"node_versions":{"comfy-core":"0.3.10"},"ue_links":[]},"version":0.4}


## Additional Context
(Please add any additional context or steps to reproduce the error here)

Other

No response

[ltdrdata]

Cannot reproduce this issue.

Update your triton to latest version

pytorch-triton==3.2.0+git4b3bb1f8
triton==3.2.0

[edisonchan]

Cannot reproduce this issue.

Update your triton to latest version

pytorch-triton==3.2.0+git4b3bb1f8
triton==3.2.0

My triton was downloaded from here: pytorch-triton==3.2.0+git4b3bb1f8, it is 3.2.0.

[ltdrdata]

triton-lang/triton#6080

[LukeG89]

Have you followed all the installation instructions in the triton-windows repo?

There are extra steps for installing it in python_embeded:

Also, I see your card is not officially supported:

[edisonchan]

Have you followed all the installation instructions in the triton-windows repo?

There are extra steps for installing it in python_embeded:

Also, I see your card is not officially supported:

I know. I have some other cards which are sm 120.

[woct0rdho]

Hmm, that triton==3.2.0+git4b3bb1f8 wheel indeed caused some confusion so now I deleted it. Although it shows 3.2.0, it's actually a pre-release that is not compatible with PyTorch 2.6

Now you can directly install Triton using pip install triton-windows, and I'll formally manage the release and pre-release versions according to PyPI's spec

For most people if you don't have Blackwell (RTX 50xx) GPU, then you need Triton 3.2 and PyTorch 2.6 . Although that torch==2.6.0+cu128.nv wheel supports CUDA 12.8 and sm120, it does not support Triton 3.3

If you have Blackwell GPU, then you need Triton 3.3 (pre-release) and PyTorch 2.7 (nightly)