How can I start the build nvdiffrast , I have already completed all the steps from the instructions

[ShockWave2048]

Can't run on Windows11+cuda (3090). Maybe somewhere there are step-by-step instructions (https://nvlabs.github.io/nvdiffrast/#windows) for assembly and testing this plug?

CUDA path exists and set in env.

Thanks on answers! Error occurred when executing CRMModelerCuda:

DLL load failed while importing nvdiffrast_plugin: Module not found.

File "F:\Store\ComfyUI\ComfyUI\execution.py", line 151, in recursive_execute
output_data, output_ui = get_output_data(obj, input_data_all)
File "F:\Store\ComfyUI\ComfyUI\execution.py", line 81, in get_output_data
return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
File "F:\Store\ComfyUI\ComfyUI\execution.py", line 74, in map_node_over_list
results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
File "F:\Store\ComfyUI\ComfyUI\custom_nodes\ComfyUI-Flowty-CRM\__init__.py", line 327, in make_model
mesh = generate3d_cuda(crm_model, rgb, ccm, device)
File "F:\Store\ComfyUI\ComfyUI\custom_nodes\ComfyUI-Flowty-CRM\crmlib\inference.py", line 97, in generate3d_cuda
glctx = dr.RasterizeCudaContext()
File "C:\Users\krami\AppData\Local\Programs\Python\Python310\lib\site-packages\nvdiffrast\torch\ops.py", line 177, in __init__
self.cpp_wrapper = _get_plugin().RasterizeCRStateWrapper(cuda_device_idx)
File "C:\Users\krami\AppData\Local\Programs\Python\Python310\lib\site-packages\nvdiffrast\torch\ops.py", line 118, in _get_plugin
torch.utils.cpp_extension.load(name=plugin_name, sources=source_paths, extra_cflags=opts, extra_cuda_cflags=opts+['-lineinfo'], extra_ldflags=ldflags, with_cuda=True, verbose=False)
File "C:\Users\krami\AppData\Local\Programs\Python\Python310\lib\site-packages\torch\utils\cpp_extension.py", line 1306, in load
return _jit_compile(
File "C:\Users\krami\AppData\Local\Programs\Python\Python310\lib\site-packages\torch\utils\cpp_extension.py", line 1736, in _jit_compile
return _import_module_from_library(name, build_directory, is_python_module)
File "C:\Users\krami\AppData\Local\Programs\Python\Python310\lib\site-packages\torch\utils\cpp_extension.py", line 2132, in _import_module_from_library
module = importlib.util.module_from_spec(spec)
File "", line 571, in module_from_spec
File "", line 1176, in create_module
File "", line 241, in _call_with_frames_removed

[flowtyone]

Please check out https://github.com/flowtyone/ComfyUI-Flowty-CRM/issues/4

[Pedroman1]

module from hell so difficult cannot get this nvdiffrast to work seriously its a tough sob

[ShockWave2048]

I'm ready to assemble the module step by step, specify it in the paths, and configure it. But I don’t see instructions anywhere on how it is compiled and installed in a Python extension! Even in the official documentation. The compilation is indicated there in passing, that's all.

[EIegance]

For anyone coming across this looking for a fix; the only thing that solved it for me after many hours was:

• Switching off of ComfyUI portable version (Probably isn't necessary as long as you make sure you're not running cu118, just something I did)
• Uninstalling Torch, and everything related to cu118
• Installing PyTorch cu112 specifically the nightly version pip3 install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/cu121
• Then reinstalling Flowty-CRM + all of its requirements

I basically reinstalled everything and then made sure ComfyUI was running PyTorch 2.4.0+cu121. I doubt you'll have to do that as long as you fully switch your PyTorch version. But in case you still can't get it working id recommend starting fresh, and its easy if you still have your _ComfyUI_windowsportable to import everything from.