How can I get these pruned GGUFs to work in ComfyUI?
All existing GGUF loaders return the error “This model is not currently supported - (Unknown model architecture!)”. How can I load these pruned models into ComfyUI?
I found the solution. https://github.com/city96/ComfyUI-GGUF/issues/471 His GGUF fork loads them: https://github.com/molbal/ComfyUI-GGUF
Passed the GGUF loader but got this error: "SamplerCustomAdvanced failed"
I'm using this workflow: https://comfy.org/workflows/a781503cf508-a781503cf508
I'm using the workflow from Rebel, and everything is working fine. https://huggingface.co/realrebelai/MiniMax-H3_GGUFs/tree/main
Apologies, that was my mistake; I had not set the pixel count as a multiple of 32. The fork works excellently.
Tried all GGUF loaders redownload text encoders.
ValueError: This gguf file is incompatible with llama.cpp!
Consider using safetensors or a compatible gguf file
(\qwen3vl_32b_minimax_h3-Q2_K_M.gguf)
Tried all GGUF loaders redownload text encoders.
ValueError: This gguf file is incompatible with llama.cpp! Consider using safetensors or a compatible gguf file (\qwen3vl_32b_minimax_h3-Q2_K_M.gguf)
You need to set it to WAN, not MiniMax. This model is even supported by the old GGUF loader.
Tried all GGUF loaders redownload text encoders.
ValueError: This gguf file is incompatible with llama.cpp! Consider using safetensors or a compatible gguf file (\qwen3vl_32b_minimax_h3-Q2_K_M.gguf)You need to set it to WAN, not MiniMax. This model is even supported by the old GGUF loader.
Does not work for me. Mine is qwen3vl_32b_minimax_h3-Q4_K_M.gguf and minimax_h3_fl2va_pruned-Q4_K.gguf
[ERROR] !!! Exception during processing !!! This gguf file is incompatible with llama.cpp!
Consider using safetensors or a compatible gguf file
(D:\soft\ComfyUIData\models\text_encoders\qwen3vl_32b_minimax_h3-Q4_K_M.gguf)
[ERROR] Traceback (most recent call last):
File "C:\Users\Administrator\ComfyUI-Installs\ComfyUI\ComfyUI\execution.py", line 545, 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\Administrator\ComfyUI-Installs\ComfyUI\ComfyUI\execution.py", line 344, 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\Administrator\ComfyUI-Installs\ComfyUI\ComfyUI\execution.py", line 318, in _async_map_node_over_list
await process_inputs(input_dict, i)
File "C:\Users\Administrator\ComfyUI-Installs\ComfyUI\ComfyUI\execution.py", line 306, in process_inputs
result = f(**inputs)
^^^^^^^^^^^
File "D:\soft\ComfyUIData\custom_nodes\ComfyUI-GGUF\nodes.py", line 251, in load_clip
return (self.load_patcher([clip_path], clip_type, self.load_data([clip_path])),)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\soft\ComfyUIData\custom_nodes\ComfyUI-GGUF\nodes.py", line 227, in load_data
sd = gguf_clip_loader(p)
^^^^^^^^^^^^^^^^^^^
File "D:\soft\ComfyUIData\custom_nodes\ComfyUI-GGUF\loader.py", line 471, in gguf_clip_loader
sd, extra = gguf_sd_loader(path, is_text_model=True)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\soft\ComfyUIData\custom_nodes\ComfyUI-GGUF\loader.py", line 98, in gguf_sd_loader
raise ValueError(f"This gguf file is incompatible with llama.cpp!\nConsider using safetensors or a compatible gguf file\n({path})")
ValueError: This gguf file is incompatible with llama.cpp!
Consider using safetensors or a compatible gguf file
(D:\soft\ComfyUIData\models\text_encoders\qwen3vl_32b_minimax_h3-Q4_K_M.gguf)
me too
[ERROR] !!! Exception during processing !!! This gguf file is incompatible with llama.cpp!
Consider using safetensors or a compatible gguf file
(D:\soft\ComfyUIData\models\text_encoders\qwen3vl_32b_minimax_h3-Q4_K_M.gguf)
[ERROR] Traceback (most recent call last):
File "C:\Users\Administrator\ComfyUI-Installs\ComfyUI\ComfyUI\execution.py", line 545, 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\Administrator\ComfyUI-Installs\ComfyUI\ComfyUI\execution.py", line 344, 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\Administrator\ComfyUI-Installs\ComfyUI\ComfyUI\execution.py", line 318, in _async_map_node_over_list
await process_inputs(input_dict, i)
File "C:\Users\Administrator\ComfyUI-Installs\ComfyUI\ComfyUI\execution.py", line 306, in process_inputs
result = f(**inputs)
^^^^^^^^^^^
File "D:\soft\ComfyUIData\custom_nodes\ComfyUI-GGUF\nodes.py", line 251, in load_clip
return (self.load_patcher([clip_path], clip_type, self.load_data([clip_path])),)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\soft\ComfyUIData\custom_nodes\ComfyUI-GGUF\nodes.py", line 227, in load_data
sd = gguf_clip_loader(p)
^^^^^^^^^^^^^^^^^^^
File "D:\soft\ComfyUIData\custom_nodes\ComfyUI-GGUF\loader.py", line 471, in gguf_clip_loader
sd, extra = gguf_sd_loader(path, is_text_model=True)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\soft\ComfyUIData\custom_nodes\ComfyUI-GGUF\loader.py", line 98, in gguf_sd_loader
raise ValueError(f"This gguf file is incompatible with llama.cpp!\nConsider using safetensors or a compatible gguf file\n({path})")
ValueError: This gguf file is incompatible with llama.cpp!
Consider using safetensors or a compatible gguf file
(D:\soft\ComfyUIData\models\text_encoders\qwen3vl_32b_minimax_h3-Q4_K_M.gguf)me too
Use vae files here https://huggingface.co/nif0/Qwen3-VL-32B-Instruct-MiniMax-H3-GGUF resolved my issue