comfyui break

[chenyongshao23]

why after install the envirement,my comfyui totally broken, could not open ,i have to reinstall

[Aero-Ex]

what issues you got?

[chenyongshao23]

i now use a totaly new comfyui,after i install all the command,the comfyui can't find the segvigen node:
(ComfyUIcys) D:\ComfyUIcys>uv pip install -r "D:\ComfyUIcys\custom_nodes\ComfyUI-SegviGen\requirements.txt"
Resolved 52 packages in 693ms
Prepared 9 packages in 5.21s
Uninstalled 9 packages in 1.30s
░░░░░░░░░░░░░░░░░░░░ [0/11] Installing wheels... warning: Failed to hardlink files; falling back to full copy. This may lead to degraded performance.
If the cache and target directories are on different filesystems, hardlinking may not be supported.
If this is intentional, set export UV_LINK_MODE=copy or use --link-mode=copy to suppress this warning.
Installed 11 packages in 1.11s

• fast-simplification==0.1.13

• huggingface-hub==1.3.5

• huggingface-hub==0.36.2

• imageio==2.37.3

• imageio==2.37.2

• numpy==2.4.1

• numpy==2.2.6

• opencv-python-headless==4.13.0.92

• opencv-python-headless==4.12.0.88

• pillow==12.1.0

• pillow==12.0.0

• rembg==2.0.73

• rembg==2.0.69

• timm==1.0.25

• timm==1.0.22

• transformers==5.0.0

• transformers==4.57.3

• trimesh==4.11.4

• trimesh==4.10.1
• zstandard==0.25.0

(ComfyUIcys) D:\ComfyUIcys>python "D:\ComfyUIcys\custom_nodes\ComfyUI-SegviGen\install.py"
--- ComfyUI-SegviGen Installation ---
D:\ComfyUIcys.venv\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]
Detected Environment:
OS: Windows
Python: cp312
PyTorch: torch210
CUDA: cu130

Installing requirements from D:\ComfyUIcys\custom_nodes\ComfyUI-SegviGen\requirements.txt...
Audited 13 packages in 24ms

--- Installing flash-attn ---
Found matching wheel: https://github.com/PozzettiAndrea/cuda-wheels/releases/download/flash_attn-latest/flash_attn-2.8.3%2Bcu130torch2.10-cp312-cp312-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl
Resolved 12 packages in 1.31s
error: Failed to determine installation plan
Caused by: A URL dependency is incompatible with the current platform: https://github.com/PozzettiAndrea/cuda-wheels/releases/download/flash_attn-latest/flash_attn-2.8.3%2Bcu130torch2.10-cp312-cp312-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl

hint: The wheel is compatible with Linux (manylinux_2_24_x86_64), Linux (manylinux_2_28_x86_64), but you're on Windows (win_amd64)
Error executing command: Command '['D:\ComfyUIcys\.venv\Scripts\python.exe', '-m', 'uv', 'pip', 'install', 'https://github.com/PozzettiAndrea/cuda-wheels/releases/download/flash_attn-latest/flash_attn-2.8.3%2Bcu130torch2.10-cp312-cp312-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl']' returned non-zero exit status 2.

--- Installing triton-windows ---
Resolved 1 package in 204ms
Audited 1 package in 0.24ms

Installing CUDA wheels...
Installing cumesh...
Found matching wheel: https://github.com/PozzettiAndrea/cuda-wheels/releases/download/cumesh-latest/cumesh-0.0.1%2Bcu130torch2.10-cp312-cp312-manylinux_2_35_x86_64.whl
Resolved 13 packages in 2.81s
error: Failed to determine installation plan
Caused by: A URL dependency is incompatible with the current platform: https://github.com/PozzettiAndrea/cuda-wheels/releases/download/cumesh-latest/cumesh-0.0.1%2Bcu130torch2.10-cp312-cp312-manylinux_2_35_x86_64.whl

hint: The wheel is compatible with Linux (manylinux_2_35_x86_64), but you're on Windows (win_amd64)
Error executing command: Command '['D:\ComfyUIcys\.venv\Scripts\python.exe', '-m', 'uv', 'pip', 'install', 'https://github.com/PozzettiAndrea/cuda-wheels/releases/download/cumesh-latest/cumesh-0.0.1%2Bcu130torch2.10-cp312-cp312-manylinux_2_35_x86_64.whl']' returned non-zero exit status 2.
Warning: Failed to install cumesh.
Installing flex-gemm...
Found matching wheel: https://github.com/PozzettiAndrea/cuda-wheels/releases/download/flex_gemm-latest/flex_gemm-1.0.0%2Bcu130torch2.10-cp312-cp312-manylinux_2_34_x86_64.manylinux_2_35_x86_64.whl
Resolved 12 packages in 1.42s
error: Failed to determine installation plan
Caused by: A URL dependency is incompatible with the current platform: https://github.com/PozzettiAndrea/cuda-wheels/releases/download/flex_gemm-latest/flex_gemm-1.0.0%2Bcu130torch2.10-cp312-cp312-manylinux_2_34_x86_64.manylinux_2_35_x86_64.whl

hint: The wheel is compatible with Linux (manylinux_2_34_x86_64), Linux (manylinux_2_35_x86_64), but you're on Windows (win_amd64)
Error executing command: Command '['D:\ComfyUIcys\.venv\Scripts\python.exe', '-m', 'uv', 'pip', 'install', 'https://github.com/PozzettiAndrea/cuda-wheels/releases/download/flex_gemm-latest/flex_gemm-1.0.0%2Bcu130torch2.10-cp312-cp312-manylinux_2_34_x86_64.manylinux_2_35_x86_64.whl']' returned non-zero exit status 2.
Warning: Failed to install flex-gemm.
Installing nvdiffrast...
Found matching wheel: https://github.com/PozzettiAndrea/cuda-wheels/releases/download/nvdiffrast-latest/nvdiffrast-0.4.0%2Bcu130torch2.10-cp312-cp312-manylinux_2_34_x86_64.manylinux_2_35_x86_64.whl
Resolved 2 packages in 3.04s
error: Failed to determine installation plan
Caused by: A URL dependency is incompatible with the current platform: https://github.com/PozzettiAndrea/cuda-wheels/releases/download/nvdiffrast-latest/nvdiffrast-0.4.0%2Bcu130torch2.10-cp312-cp312-manylinux_2_34_x86_64.manylinux_2_35_x86_64.whl

hint: The wheel is compatible with Linux (manylinux_2_34_x86_64), Linux (manylinux_2_35_x86_64), but you're on Windows (win_amd64)
Error executing command: Command '['D:\ComfyUIcys\.venv\Scripts\python.exe', '-m', 'uv', 'pip', 'install', 'https://github.com/PozzettiAndrea/cuda-wheels/releases/download/nvdiffrast-latest/nvdiffrast-0.4.0%2Bcu130torch2.10-cp312-cp312-manylinux_2_34_x86_64.manylinux_2_35_x86_64.whl']' returned non-zero exit status 2.
Warning: Failed to install nvdiffrast.
Installing nvdiffrec-render...
Found matching wheel: https://github.com/PozzettiAndrea/cuda-wheels/releases/download/nvdiffrec_render-latest/nvdiffrec_render-0.0.1%2Bcu130torch2.10-cp312-cp312-manylinux_2_34_x86_64.manylinux_2_35_x86_64.whl
Resolved 12 packages in 1.91s
error: Failed to determine installation plan
Caused by: A URL dependency is incompatible with the current platform: https://github.com/PozzettiAndrea/cuda-wheels/releases/download/nvdiffrec_render-latest/nvdiffrec_render-0.0.1%2Bcu130torch2.10-cp312-cp312-manylinux_2_34_x86_64.manylinux_2_35_x86_64.whl

hint: The wheel is compatible with Linux (manylinux_2_34_x86_64), Linux (manylinux_2_35_x86_64), but you're on Windows (win_amd64)
Error executing command: Command '['D:\ComfyUIcys\.venv\Scripts\python.exe', '-m', 'uv', 'pip', 'install', 'https://github.com/PozzettiAndrea/cuda-wheels/releases/download/nvdiffrec_render-latest/nvdiffrec_render-0.0.1%2Bcu130torch2.10-cp312-cp312-manylinux_2_34_x86_64.manylinux_2_35_x86_64.whl']' returned non-zero exit status 2.
Warning: Failed to install nvdiffrec-render.
Installing o-voxel...
Found matching wheel: https://github.com/PozzettiAndrea/cuda-wheels/releases/download/o_voxel-latest/o_voxel-0.0.1%2Bcu130torch2.10-cp312-cp312-manylinux_2_34_x86_64.manylinux_2_35_x86_64.whl
x No solution found when resolving dependencies:
`-> Because flex-gemm was not found in the package registry and o-voxel==0.0.1+cu130torch2.10 depends on flex-gemm,
we can conclude that o-voxel==0.0.1+cu130torch2.10 cannot be used.
And because only o-voxel==0.0.1+cu130torch2.10 is available and you require o-voxel, we can conclude that your
requirements are unsatisfiable.
Error executing command: Command '['D:\ComfyUIcys\.venv\Scripts\python.exe', '-m', 'uv', 'pip', 'install', 'https://github.com/PozzettiAndrea/cuda-wheels/releases/download/o_voxel-latest/o_voxel-0.0.1%2Bcu130torch2.10-cp312-cp312-manylinux_2_34_x86_64.manylinux_2_35_x86_64.whl']' returned non-zero exit status 1.
Warning: Failed to install o-voxel.

==================================================
RECOMMENDED ENVIRONMENTS

If you encounter 404 errors, it is likely because your environment
combination is not yet supported on the wheel repository.

Preferred configurations:

1. CUDA 12.4 + PyTorch 2.5 (Most compatible)
2. CUDA 12.6 + PyTorch 2.6 (Latest compatible)
3. CUDA 12.1 + PyTorch 2.4 (Legacy compatible)

You can install the preferred Torch version using:
pip install torch==2.5.1 --index-url https://download.pytorch.org/whl/cu124

Installation complete!

(ComfyUIcys) D:\ComfyUIcys>

[chenyongshao23]

example workflow could not open

[Aero-Ex]

i have fixed it install again

[chenyongshao23]

now is works,i can see the node in canvas ，and successfuly run the workflow,but the workflow show error:
ValueError: Unsupported mesh type for baking: <class 'comfy_api.latest._util.geometry_types.File3D'>. Available attributes: ['class', 'delattr', 'dict', 'dir', 'doc', 'eq', 'format', 'ge', 'getattribute', 'getstate', 'gt', 'hash', 'init', 'init_subclass', 'le', 'lt', 'module', 'ne', 'new', 'reduce', 'reduce_ex', 'repr', 'setattr', 'sizeof', 'str', 'subclasshook', 'weakref', '_format', '_infer_format', '_source', 'format', 'get_bytes', 'get_data', 'get_source', 'is_disk_backed', 'save_to']

File "C:\Users\TU-CYS\AppData\Local\Programs\ComfyUI\resources\ComfyUI\execution.py", line 525, 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, v3_data=v3_data)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "C:\Users\TU-CYS\AppData\Local\Programs\ComfyUI\resources\ComfyUI\execution.py", line 334, 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, v3_data=v3_data)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "C:\Users\TU-CYS\AppData\Local\Programs\ComfyUI\resources\ComfyUI\execution.py", line 308, in _async_map_node_over_list
await process_inputs(input_dict, i)

File "C:\Users\TU-CYS\AppData\Local\Programs\ComfyUI\resources\ComfyUI\execution.py", line 296, in process_inputs
result = f(**inputs)
^^^^^^^^^^^

File "D:\ComfyUIcys\custom_nodes\ComfyUI-SegviGen\nodes.py", line 339, in bake
raise ValueError(f"Unsupported mesh type for baking: {type(mesh)}. Available attributes: {dir(mesh)}")

[chenyongshao23]

[chenyongshao23]

i use defual basket, bread mesh expample

[Aero-Ex]

use mesh_path output from load3d and animation node instead of model_3d

[chenyongshao23]

now all it work，thanks very much!
i have a question,if i use other mesh ,how can i generate the 2D guid segment image using AI MODEL

[Aero-Ex]

you can use Sam3 segmentation https://github.com/PozzettiAndrea/ComfyUI-SAM3 this to create segmentation mask

[chenyongshao23]

then what prompt do you use to generate all colored flat segment image so nice in sam3 at once,i can only mask one of the object's segment

[Aero-Ex]

SegvGen_PixelArtistry_test.json use this workflow from pixel artistry

[chenyongshao23]

thanks you very much again！

[chenyongshao23]

is seems nodes in Segvgen_PixelArtistry_test.json are newer?

[Aero-Ex]

they are old replace them with new nodes