Stucked at MultiViews

[Milor123]

I've check some issues similars like #57

What things tried?

Manually compiled both files just in case; I had also tried the precompiled ones.
but I don't know what's happening, it just stays stuck there, not consuming resources, doing nothing

Total VRAM 12282 MB, total RAM 48445 MB
pytorch version: 2.8.0+cu128
xformers version: 0.0.32.post2
Set vram state to: LOW_VRAM
Device: cuda:0 NVIDIA GeForce RTX 4070 : native
Using sage attention
Python version: 3.12.10 (tags/v3.12.10:0cc8128, Apr 8 2025, 12:21:36) [MSC v.1943 64 bit (AMD64)]
ComfyUI version: 0.3.67
ComfyUI frontend version: 1.28.8

Windows 11 24h2, using comfy-cli version + UV

what should i do guys?

Non GPU usage here:

Compile manually whl in the UV venv of my comfyui

User   (.venv)  custom_nodes  ➜ (master  1)  ♥ 17:52  uv pip install .\ComfyUI-Hunyuan3d-2-1\hy3dpaint\custom_rasterizer\ --no-build-isolation
Using Python 3.12.10 environment at: C:\Users\User\Documents\TEO\ComfyUI\.venv
Resolved 1 package in 2.88s
      Built custom-rasterizer @ file:///C:/Users/User/Documents/TEO/ComfyUI/custom_nodes/ComfyUI-Hunyuan3d-2-1/hy3dpaint/custom_rasterizer
Prepared 1 package in 53.19s
Uninstalled 1 package in 4ms
Installed 1 package in 9ms
 - custom-rasterizer==0.1 (from file:///C:/Users/User/Documents/TEO/ComfyUI/custom_nodes/ComfyUI-Hunyuan3d-2-1/hy3dpaint/custom_rasterizer/dist/custom_rasterizer-0.1-cp312-cp312-win_amd64.whl)
 + custom-rasterizer==0.1 (from file:///C:/Users/User/Documents/TEO/ComfyUI/custom_nodes/ComfyUI-Hunyuan3d-2-1/hy3dpaint/custom_rasterizer)
User   (.venv)  custom_nodes  ➜ (master  1)  ♥ 17:53  uv pip install .\ComfyUI-Hunyuan3d-2-1\hy3dpaint\DifferentiableRenderer --no-build-isolation
Using Python 3.12.10 environment at: C:\Users\User\Documents\TEO\ComfyUI\.venv
Resolved 1 package in 661ms
      Built mesh-inpaint-processor @ file:///C:/Users/User/Documents/TEO/ComfyUI/custom_nodes/ComfyUI-Hunyuan3d-2-1/hy3dpaint/DifferentiableRenderer
Prepared 1 package in 5.59s
Uninstalled 1 package in 2ms
Installed 1 package in 8ms                                                                                                                                                                                      
 - mesh-inpaint-processor==0.0.0 (from file:///C:/Users/User/Documents/TEO/ComfyUI/custom_nodes/ComfyUI-Hunyuan3d-2-1/hy3dpaint/DifferentiableRenderer/dist/mesh_inpaint_processor-0.0.0-cp312-cp312-win_amd64.whl)
 + mesh-inpaint-processor==0.0.0 (from file:///C:/Users/User/Documents/TEO/ComfyUI/custom_nodes/ComfyUI-Hunyuan3d-2-1/hy3dpaint/DifferentiableRenderer)

[zdaar]

I got the same issue

got prompt
Megapixel resize: 1502x1613 to 960x1024 (Max area: 1000000 pixels)
using moe
using moe
using moe
using moe
using moe
using moe
PointCrossAttentionEncoder INFO: pc_sharpedge_size is not given, using pc_size as pc_sharpedge_size
Diffusion Sampling:: 100%|█████████████████████████████████| 25/25 [00:17<00:00,  1.41it/s]
PointCrossAttentionEncoder INFO: pc_sharpedge_size is not given, using pc_size as pc_sharpedge_size
2025-11-09 19:21:05,261 - hy3dgen.shapgen - INFO - FlashVDMVolumeDecoding Resolution: [63, 126, 252]
2025-11-09 19:21:05,261 - hy3dgen.shapgen - INFO - FlashVDMVolumeDecoding Resolution: [63, 126, 252]
FlashVDMVolumeDecoding Resolution: [63, 126, 252]
FlashVDM Volume Decoding: 100%|██████████████████████████████| 5/5 [00:00<00:00, 44.97it/s]
Decoded mesh with 394337 vertices and 1496574 faces
Removed floaters, resulting in 394201 vertices and 788438 faces
Removed degenerate faces, resulting in 394201 vertices and 788438 faces
Reduced faces, resulting in 99982 vertices and 200000 faces

[masqueradeljb]

I got the same issue, but my problem was it stuck at multiViews generator. It turned out to be an issue importing the custom_rasterizer. Downgrading to cuda 12.6 fixed the issue.

[mc-maloney]

Same issue here. I'm on CUDA 12.4, got the correct rasterizers installed, etc. For me it's hanging on the Hunyuan 3D 2.1 MultiViews Generator node, not the MeshUVWrap one. My log looks the same as the ones posted above, just ends with the "reduced faces" line and then nothing.

[jyhxym]

一样的参考之后CUDA 12.8降级到12.6后，这个节点通过了