AttributeError: 'Hunyuan3DDiTFlowMatchingPipeline' object has no attribute 'scheduler'

[TeKett]

First time using, so im just trying out the existing workflows to make sure things work and to learn how to use this. So i used the full workflow for Hunyuan.

File "/home/user/venvs/ComfyUI/execution.py", line 303, in _async_map_node_over_list
    await process_inputs(input_dict, i)
  File "/home/user/venvs/ComfyUI/execution.py", line 291, in process_inputs
    result = f(**inputs)
             ^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/venv/lib/python3.12/site-packages/torch/utils/_contextlib.py", line 116, in decorate_context
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/custom_nodes/ComfyUI-3D-Pack/nodes.py", line 5432, in generate
    outputs = shapegen_pipe(
              ^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/venv/lib/python3.12/site-packages/torch/utils/_contextlib.py", line 116, in decorate_context
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/custom_nodes/ComfyUI-3D-Pack/Gen_3D_Modules/Hunyuan3D_2_1/hy3dshape/hy3dshape/pipelines.py", line 740, in __call__
    self.scheduler,
    ^^^^^^^^^^^^^^
AttributeError: 'Hunyuan3DDiTFlowMatchingPipeline' object has no attribute 'scheduler'

[TeKett]

Also whould it be possible for you to add some feedback for the user for the things that the nodes are doing? The only feedback im getting are some warnings, thats the only way i know anything is happening at all.

[TeKett]

So i noticed the other issue, with the same topic, sorry. It was an issue on the second try, but restarting just leads into more issues.

File "/home/user/venvs/ComfyUI/execution.py", line 303, in _async_map_node_over_list
    await process_inputs(input_dict, i)
  File "/home/user/venvs/ComfyUI/execution.py", line 291, in process_inputs
    result = f(**inputs)
             ^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/venv/lib/python3.12/site-packages/torch/utils/_contextlib.py", line 116, in decorate_context
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/custom_nodes/ComfyUI-3D-Pack/nodes.py", line 5509, in generate
    result_path = texgen_pipe(
                  ^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/venv/lib/python3.12/site-packages/torch/utils/_contextlib.py", line 116, in decorate_context
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/custom_nodes/ComfyUI-3D-Pack/Gen_3D_Modules/Hunyuan3D_2_1/hy3dpaint/textureGenPipeline.py", line 184, in __call__
    texture = self.view_processor.texture_inpaint(texture, mask_np)
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/custom_nodes/ComfyUI-3D-Pack/Gen_3D_Modules/Hunyuan3D_2_1/hy3dpaint/utils/pipeline_utils.py", line 132, in texture_inpaint
    texture_np = self.render.uv_inpaint(texture, mask)
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/venv/lib/python3.12/site-packages/torch/utils/_contextlib.py", line 116, in decorate_context
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/custom_nodes/ComfyUI-3D-Pack/Gen_3D_Modules/Hunyuan3D_2_1/hy3dpaint/DifferentiableRenderer/MeshRender.py", line 1408, in uv_inpaint
    texture_np, mask = meshVerticeInpaint(texture_np, mask, vtx_pos, vtx_uv, pos_idx, uv_idx)
                       ^^^^^^^^^^^^^^^^^^
NameError: name 'meshVerticeInpaint' is not defined

[DenisKochetov]

@TeKett Hi, this means the wheel wasn’t installed correctly.

Please go to the following folder:
ComfyUI-3D-Pack/Gen_3D_Modules/Hunyuan3D_2_1/hy3dpaint/DifferentiableRenderer

Then run the appropriate script depending on your OS:

• compile_mesh_painter.bat (Windows)
• compile_mesh_painter.sh (Linux / macOS)

This will build the required library.

[TeKett]

Well it did not fail the install py script, there was no errors. but running compile_mesh_painter.sh fixed it.

To fix the first issue with the scheduler, i downloaded the dev branch. But im getting a new issue on the second job instead. But it only seems to be an issue on Hunyuan, i tried partcrafter and it does not break on the second job.

!!! Exception during processing !!! can't convert cuda:0 device type tensor to numpy. Use Tensor.cpu() to copy the tensor to host memory first.
Traceback (most recent call last):
  File "/home/user/venvs/ComfyUI/execution.py", line 518, 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 "/home/user/venvs/ComfyUI/execution.py", line 329, 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 "/home/user/venvs/ComfyUI/execution.py", line 303, in _async_map_node_over_list
    await process_inputs(input_dict, i)
  File "/home/user/venvs/ComfyUI/execution.py", line 291, in process_inputs
    result = f(**inputs)
             ^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/venv/lib/python3.12/site-packages/torch/utils/_contextlib.py", line 116, in decorate_context
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/custom_nodes/ComfyUI-3D-Pack/nodes.py", line 5522, in generate
    result_path = texgen_pipe(
                  ^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/venv/lib/python3.12/site-packages/torch/utils/_contextlib.py", line 116, in decorate_context
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/custom_nodes/ComfyUI-3D-Pack/Gen_3D_Modules/Hunyuan3D_2_1/hy3dpaint/textureGenPipeline.py", line 152, in __call__
    multiviews_pbr = self.models["multiview_model"](
                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/venv/lib/python3.12/site-packages/torch/utils/_contextlib.py", line 116, in decorate_context
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/custom_nodes/ComfyUI-3D-Pack/Gen_3D_Modules/Hunyuan3D_2_1/hy3dpaint/utils/multiview_utils.py", line 68, in __call__
    pils = self.forward_one(
           ^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/custom_nodes/ComfyUI-3D-Pack/Gen_3D_Modules/Hunyuan3D_2_1/hy3dpaint/utils/multiview_utils.py", line 113, in forward_one
    mvd_image = self.pipeline(
                ^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/venv/lib/python3.12/site-packages/torch/utils/_contextlib.py", line 116, in decorate_context
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/.cache/huggingface/modules/diffusers_modules/local/pipeline.py", line 329, in __call__
    images = self.denoise(
             ^^^^^^^^^^^^^
  File "/home/user/.cache/huggingface/modules/diffusers_modules/local/pipeline.py", line 590, in denoise
    timesteps, num_inference_steps = retrieve_timesteps(
                                     ^^^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/venv/lib/python3.12/site-packages/diffusers/pipelines/stable_diffusion/pipeline_stable_diffusion.py", line 149, in retrieve_timesteps
    scheduler.set_timesteps(num_inference_steps, device=device, **kwargs)
  File "/home/user/venvs/ComfyUI/venv/lib/python3.12/site-packages/diffusers/schedulers/scheduling_unipc_multistep.py", line 348, in set_timesteps
    sigmas = np.array(((1 - self.alphas_cumprod) / self.alphas_cumprod) ** 0.5)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/venv/lib/python3.12/site-packages/torch/_tensor.py", line 1223, in __array__
    return handle_torch_function(Tensor.__array__, (self,), self, dtype=dtype)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/venv/lib/python3.12/site-packages/torch/overrides.py", line 1721, in handle_torch_function
    result = mode.__torch_function__(public_api, types, args, kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/venv/lib/python3.12/site-packages/torch/utils/_device.py", line 104, in __torch_function__
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/venv/lib/python3.12/site-packages/torch/_tensor.py", line 1225, in __array__
    return self.numpy()
           ^^^^^^^^^^^^
TypeError: can't convert cuda:0 device type tensor to numpy. Use Tensor.cpu() to copy the tensor to host memory first.

[TeKett]

I also tested out TripoSG and TripoSR, but both fails since the mask is created with a resolution of 64x64 rather then copying the image resolution, so it always mismatches unless the image is 64x64. However, giving it a 64x64 image and getting past that stage with TripoSG i get
OSError: Error no file named diffusion_pytorch_model.bin found in directory /home/user/venvs/ComfyUI/custom_nodes/ComfyUI-3D-Pack/Checkpoints/Diffusers/VAST-AI/TripoSG/transformer.

[TeKett]

And Trilles gives

Traceback (most recent call last):
  File "/home/user/venvs/ComfyUI/execution.py", line 518, 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 "/home/user/venvs/ComfyUI/execution.py", line 329, 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 "/home/user/venvs/ComfyUI/execution.py", line 303, in _async_map_node_over_list
    await process_inputs(input_dict, i)
  File "/home/user/venvs/ComfyUI/execution.py", line 291, in process_inputs
    result = f(**inputs)
             ^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/custom_nodes/ComfyUI-3D-Pack/nodes.py", line 4105, in load_pipe
    pipe = TrellisImageTo3DPipeline.from_pretrained(repo_id)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/custom_nodes/ComfyUI-3D-Pack/Gen_3D_Modules/TRELLIS/trellis/pipelines/trellis_image_to_3d.py", line 54, in from_pretrained
    pipeline = super(TrellisImageTo3DPipeline, TrellisImageTo3DPipeline).from_pretrained(path)
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/custom_nodes/ComfyUI-3D-Pack/Gen_3D_Modules/TRELLIS/trellis/pipelines/base.py", line 40, in from_pretrained
    k: models.from_pretrained(f"{path}/{v}")
       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/custom_nodes/ComfyUI-3D-Pack/Gen_3D_Modules/TRELLIS/trellis/models/__init__.py", line 59, in from_pretrained
    model = __getattr__(config['name'])(**config['args'], **kwargs)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/custom_nodes/ComfyUI-3D-Pack/Gen_3D_Modules/TRELLIS/trellis/models/structured_latent_vae/decoder_mesh.py", line 108, in __init__
    SparseSubdivideBlock3d(
  File "/home/user/venvs/ComfyUI/custom_nodes/ComfyUI-3D-Pack/Gen_3D_Modules/TRELLIS/trellis/models/structured_latent_vae/decoder_mesh.py", line 43, in __init__
    sp.SparseConv3d(channels, self.out_channels, 3, indice_key=f"res_{self.out_resolution}"),
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/custom_nodes/ComfyUI-3D-Pack/Gen_3D_Modules/TRELLIS/trellis/modules/sparse/conv/conv_spconv.py", line 11, in __init__
    import spconv.pytorch as spconv
  File "/home/user/venvs/ComfyUI/venv/lib/python3.12/site-packages/spconv/pytorch/__init__.py", line 7, in <module>
    from spconv.pytorch.core import SparseConvTensor
  File "/home/user/venvs/ComfyUI/venv/lib/python3.12/site-packages/spconv/pytorch/core.py", line 21, in <module>
    from spconv.tools import CUDAKernelTimer
  File "/home/user/venvs/ComfyUI/venv/lib/python3.12/site-packages/spconv/tools.py", line 16, in <module>
    from spconv.cppconstants import CPU_ONLY_BUILD
  File "/home/user/venvs/ComfyUI/venv/lib/python3.12/site-packages/spconv/cppconstants.py", line 15, in <module>
    import spconv.core_cc as _ext
ImportError: arg(): could not convert default argument 'workspace: tv::Tensor' in method '<class 'spconv.core_cc.csrc.sparse.convops.gemmops.GemmTunerSimple'>.run_with_tuned_result' into a Python object (type not registered yet?)

‎
Trying to run Trellis a second time gives this, but could be related to the premature exit from the error above.

Traceback (most recent call last):
  File "/home/user/venvs/ComfyUI/execution.py", line 518, 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 "/home/user/venvs/ComfyUI/execution.py", line 329, 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 "/home/user/venvs/ComfyUI/execution.py", line 303, in _async_map_node_over_list
    await process_inputs(input_dict, i)
  File "/home/user/venvs/ComfyUI/execution.py", line 291, in process_inputs
    result = f(**inputs)
             ^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/custom_nodes/ComfyUI-3D-Pack/nodes.py", line 4105, in load_pipe
    pipe = TrellisImageTo3DPipeline.from_pretrained(repo_id)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/custom_nodes/ComfyUI-3D-Pack/Gen_3D_Modules/TRELLIS/trellis/pipelines/trellis_image_to_3d.py", line 54, in from_pretrained
    pipeline = super(TrellisImageTo3DPipeline, TrellisImageTo3DPipeline).from_pretrained(path)
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/custom_nodes/ComfyUI-3D-Pack/Gen_3D_Modules/TRELLIS/trellis/pipelines/base.py", line 40, in from_pretrained
    k: models.from_pretrained(f"{path}/{v}")
       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/custom_nodes/ComfyUI-3D-Pack/Gen_3D_Modules/TRELLIS/trellis/models/__init__.py", line 59, in from_pretrained
    model = __getattr__(config['name'])(**config['args'], **kwargs)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/custom_nodes/ComfyUI-3D-Pack/Gen_3D_Modules/TRELLIS/trellis/models/structured_latent_vae/decoder_mesh.py", line 108, in __init__
    SparseSubdivideBlock3d(
  File "/home/user/venvs/ComfyUI/custom_nodes/ComfyUI-3D-Pack/Gen_3D_Modules/TRELLIS/trellis/models/structured_latent_vae/decoder_mesh.py", line 43, in __init__
    sp.SparseConv3d(channels, self.out_channels, 3, indice_key=f"res_{self.out_resolution}"),
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/custom_nodes/ComfyUI-3D-Pack/Gen_3D_Modules/TRELLIS/trellis/modules/sparse/conv/conv_spconv.py", line 11, in __init__
    import spconv.pytorch as spconv
  File "/home/user/venvs/ComfyUI/venv/lib/python3.12/site-packages/spconv/pytorch/__init__.py", line 7, in <module>
    from spconv.pytorch.core import SparseConvTensor
  File "/home/user/venvs/ComfyUI/venv/lib/python3.12/site-packages/spconv/pytorch/core.py", line 21, in <module>
    from spconv.tools import CUDAKernelTimer
  File "/home/user/venvs/ComfyUI/venv/lib/python3.12/site-packages/spconv/tools.py", line 16, in <module>
    from spconv.cppconstants import CPU_ONLY_BUILD
  File "/home/user/venvs/ComfyUI/venv/lib/python3.12/site-packages/spconv/cppconstants.py", line 15, in <module>
    import spconv.core_cc as _ext
ImportError: generic_type: cannot initialize type "ExternalAllocator": an object with that name is already defined

[DenisKochetov]

@TeKett thanks a lot for taking the time to test and for the detailed report — really appreciate it

Could you please try the dev branch as well? I’m pretty sure the Hunyuan issues you’re seeing are already fixed there — I was just waiting for mrforexample to merge the changes.

Important note: it would be great if you could re-run the installation from a fresh environment when testing dev, just to avoid any leftovers from the previous setup.

[TeKett]

I reinstalled from scratch with dev, and it still only works on the first job and it fails the Hunyuan TexGen loading with

!!! Exception during processing !!! cannot pickle 'Stream' object
Traceback (most recent call last):
  File "/home/user/venvs/ComfyUI/execution.py", line 518, 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 "/home/user/venvs/ComfyUI/execution.py", line 329, 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 "/home/user/venvs/ComfyUI/execution.py", line 303, in _async_map_node_over_list
    await process_inputs(input_dict, i)
  File "/home/user/venvs/ComfyUI/execution.py", line 291, in process_inputs
    result = f(**inputs)
             ^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/custom_nodes/ComfyUI-3D-Pack/nodes.py", line 5405, in load
    pipeline = Hunyuan3DPaintPipeline_2_1(conf)
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/custom_nodes/ComfyUI-3D-Pack/Gen_3D_Modules/Hunyuan3D_2_1/hy3dpaint/textureGenPipeline.py", line 88, in __init__
    self.load_models()
  File "/home/user/venvs/ComfyUI/custom_nodes/ComfyUI-3D-Pack/Gen_3D_Modules/Hunyuan3D_2_1/hy3dpaint/textureGenPipeline.py", line 93, in load_models
    self.models["multiview_model"] = multiviewDiffusionNet(self.config)
                                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/custom_nodes/ComfyUI-3D-Pack/Gen_3D_Modules/Hunyuan3D_2_1/hy3dpaint/utils/multiview_utils.py", line 43, in __init__
    pipeline = DiffusionPipeline.from_pretrained(
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/venv/lib/python3.12/site-packages/huggingface_hub/utils/_validators.py", line 114, in _inner_fn
    return fn(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/venv/lib/python3.12/site-packages/diffusers/pipelines/pipeline_utils.py", line 896, in from_pretrained
    loaded_sub_model = load_sub_model(
                       ^^^^^^^^^^^^^^^
  File "/home/user/venvs/ComfyUI/venv/lib/python3.12/site-packages/diffusers/pipelines/pipeline_loading_utils.py", line 704, in load_sub_model
    loaded_sub_model = load_method(os.path.join(cached_folder, name), **loading_kwargs)
                       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/.cache/huggingface/modules/diffusers_modules/local/modules.py", line 817, in from_pretrained
    unet_2p5d = UNet2p5DConditionModel(unet)
                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/.cache/huggingface/modules/diffusers_modules/local/modules.py", line 796, in __init__
    self.unet_dual = copy.deepcopy(unet)
                     ^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/copy.py", line 162, in deepcopy
    y = _reconstruct(x, memo, *rv)
        ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/copy.py", line 259, in _reconstruct
    state = deepcopy(state, memo)
            ^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/copy.py", line 136, in deepcopy
    y = copier(x, memo)
        ^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/copy.py", line 221, in _deepcopy_dict
    y[deepcopy(key, memo)] = deepcopy(value, memo)
                             ^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/copy.py", line 136, in deepcopy
    y = copier(x, memo)
        ^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/copy.py", line 221, in _deepcopy_dict
    y[deepcopy(key, memo)] = deepcopy(value, memo)
                             ^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/copy.py", line 162, in deepcopy
    y = _reconstruct(x, memo, *rv)
        ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/copy.py", line 259, in _reconstruct
    state = deepcopy(state, memo)
            ^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/copy.py", line 136, in deepcopy
    y = copier(x, memo)
        ^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/copy.py", line 221, in _deepcopy_dict
    y[deepcopy(key, memo)] = deepcopy(value, memo)
                             ^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/copy.py", line 136, in deepcopy
    y = copier(x, memo)
        ^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/copy.py", line 221, in _deepcopy_dict
    y[deepcopy(key, memo)] = deepcopy(value, memo)
                             ^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/copy.py", line 162, in deepcopy
    y = _reconstruct(x, memo, *rv)
        ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/copy.py", line 259, in _reconstruct
    state = deepcopy(state, memo)
            ^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/copy.py", line 136, in deepcopy
    y = copier(x, memo)
        ^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/copy.py", line 221, in _deepcopy_dict
    y[deepcopy(key, memo)] = deepcopy(value, memo)
                             ^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/copy.py", line 162, in deepcopy
    y = _reconstruct(x, memo, *rv)
        ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/copy.py", line 259, in _reconstruct
    state = deepcopy(state, memo)
            ^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/copy.py", line 136, in deepcopy
    y = copier(x, memo)
        ^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/copy.py", line 221, in _deepcopy_dict
    y[deepcopy(key, memo)] = deepcopy(value, memo)
                             ^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/copy.py", line 151, in deepcopy
    rv = reductor(4)
         ^^^^^^^^^^^
TypeError: cannot pickle 'Stream' object

[DenisKochetov]

@TeKett try unabling mmgp offloading please.
also disable option auto_cleanup, it may help

[TeKett]

disabling auto_cleanup did not help
how do i unable mmgp offloading? Is it "enable_mmgp". Unchecking this however did not help

Anyways im probably going to just keep restarting comfy every time, since i noticed the nodes are not clearing the vram after each use, like i can run the default of octree resolution of 256 and texture resolution of 768. But anything higher doesnt work, unless, i split the mesh generation and the texture generation into their own workflows, and restart comfy between the two in order to reduce the vram. Then i can do octree 512 and texture 1024.