installation problem with cuda triton eager

[tomaszjacek]

hi,

here is my environment
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 576.88 Driver Version: 576.88 CUDA Version: 12.9 |
|-----------------------------------------+------------------------+----------------------+

an here is my log

ComfyUI-Manager: installing dependencies done.

[2026-02-14 01:17:52.673] ** ComfyUI startup time: 2026-02-14 01:17:52.673
[2026-02-14 01:17:52.673] ** Platform: Windows
[2026-02-14 01:17:52.673] ** Python version: 3.13.11 (tags/v3.13.11:6278944, Dec 5 2025, 16:26:58) [MSC v.1944 64 bit (AMD64)]
[2026-02-14 01:17:52.673] ** Python executable: C:\Users\tomas\work\3d-guided-genai-rtx-bug_fix\ComfyUI_windows_portable\python_embeded\python.exe
[2026-02-14 01:17:52.673] ** ComfyUI Path: C:\Users\tomas\work\3d-guided-genai-rtx-bug_fix\ComfyUI_windows_portable\ComfyUI
[2026-02-14 01:17:52.673] ** ComfyUI Base Folder Path: C:\Users\tomas\work\3d-guided-genai-rtx-bug_fix\ComfyUI_windows_portable\ComfyUI
[2026-02-14 01:17:52.674] ** User directory: C:\Users\tomas\work\3d-guided-genai-rtx-bug_fix\ComfyUI_windows_portable\ComfyUI\user
[2026-02-14 01:17:52.674] ** ComfyUI-Manager config path: C:\Users\tomas\work\3d-guided-genai-rtx-bug_fix\ComfyUI_windows_portable\ComfyUI\user__manager\config.ini
[2026-02-14 01:17:52.674] ** Log path: C:\Users\tomas\work\3d-guided-genai-rtx-bug_fix\ComfyUI_windows_portable\ComfyUI\user\comfyui.log

Prestartup times for custom nodes:
[2026-02-14 01:17:53.519] 2.2 seconds: C:\Users\tomas\work\3d-guided-genai-rtx-bug_fix\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Manager
[2026-02-14 01:17:53.519]
[2026-02-14 01:18:03.095] C:\Users\tomas\work\3d-guided-genai-rtx-bug_fix\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\cuda_init_.py:65: FutureWarning: The pynvml package is deprecated. Please install nvidia-ml-py instead. If you did not install pynvml directly, please report this to the maintainers of the package that installed pynvml for you.
import pynvml # type: ignore[import]
[2026-02-14 01:18:04.098] Checkpoint files will always be loaded safely.
[2026-02-14 01:18:04.281] C:\Users\tomas\work\3d-guided-genai-rtx-bug_fix\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\cuda_init_.py:184: UserWarning: cudaGetDeviceCount() returned cudaErrorNotSupported, likely using older driver or on CPU machine (Triggered internally at C:\actions-runner_work\pytorch\pytorch\pytorch\c10\cuda\CUDAFunctions.cpp:88.)
return torch._C._cuda_getDeviceCount() > 0
[2026-02-14 01:18:05.470] Found comfy_kitchen backend cuda: {'available': False, 'disabled': False, 'unavailable_reason': 'CUDA not available on this system', 'capabilities': []}
[2026-02-14 01:18:05.470] Found comfy_kitchen backend triton: {'available': False, 'disabled': True, 'unavailable_reason': "ImportError: No module named 'triton'", 'capabilities': []}
[2026-02-14 01:18:05.470] Found comfy_kitchen backend eager: {'available': True, 'disabled': False, 'unavailable_reason': None, 'capabilities': ['apply_rope', 'apply_rope1', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'scaled_mm_nvfp4']}

[jpenningCA]

@tomaszjacek can you share your system specifications?