cannot access local variable 'self_attn_func' where it is not associated with a value #17
Description
ComfyUI Error Report
Error Details
- Node ID: 1878
- Node Type: LTXQ8Patch
- Exception Type: UnboundLocalError
- Exception Message: cannot access local variable 'self_attn_func' where it is not associated with a value
Stack Trace
File "E:\workspace\ComfyUI\execution.py", line 361, in execute
output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\workspace\ComfyUI\execution.py", line 236, in get_output_data
return_values = _map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\workspace\ComfyUI\execution.py", line 208, in _map_node_over_list
process_inputs(input_dict, i)
File "E:\workspace\ComfyUI\execution.py", line 197, in process_inputs
results.append(getattr(obj, func)(**inputs))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\workspace\ComfyUI\custom_nodes\ComfyUI-LTXVideo\q8_nodes.py", line 62, in patch
patcher(transformer, use_fp8_attention, True)
File "E:\workspace\ComfyUI\venv312\Lib\site-packages\q8_kernels\integration\patch_transformer.py", line 26, in patch_comfyui_native_transformer
attn_forward = cross_attn_forward(use_fp8_attention)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\workspace\ComfyUI\venv312\Lib\site-packages\q8_kernels\integration\comfyui_native.py", line 66, in cross_attn_forward
self_attn, cross_attn, is_out_tuple = get_attention_func(use_fp8_attention)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\workspace\ComfyUI\venv312\Lib\site-packages\q8_kernels\integration\utils.py", line 104, in get_attention_func
(self_attn_func, self_attn_memory_layout),
^^^^^^^^^^^^^^
System Information
- ComfyUI Version: 0.3.41
- Arguments: main.py
- OS: nt
- Python Version: 3.12.0 (tags/v3.12.0:0fb18b0, Oct 2 2023, 13:03:39) [MSC v.1935 64 bit (AMD64)]
- Embedded Python: false
- PyTorch Version: 2.7.1+cu128
Devices
- Name: cuda:0 NVIDIA GeForce RTX 3060 Laptop GPU : cudaMallocAsync
- Type: cuda
- VRAM Total: 6441926656
- VRAM Free: 5370806272
- Torch VRAM Total: 0
- Torch VRAM Free: 0
Logs
2025-06-28T17:50:01.912219 - [START] Security scan2025-06-28T17:50:01.912219 -
2025-06-28T17:50:02.883446 - [DONE] Security scan2025-06-28T17:50:02.883446 -
2025-06-28T17:50:03.012652 - ## ComfyUI-Manager: installing dependencies done.2025-06-28T17:50:03.012652 -
2025-06-28T17:50:03.012652 - ** ComfyUI startup time:2025-06-28T17:50:03.012652 - 2025-06-28T17:50:03.012652 - 2025-06-28 17:50:03.0122025-06-28T17:50:03.012652 -
2025-06-28T17:50:03.012652 - ** Platform:2025-06-28T17:50:03.012652 - 2025-06-28T17:50:03.012652 - Windows2025-06-28T17:50:03.012652 -
2025-06-28T17:50:03.012652 - ** Python version:2025-06-28T17:50:03.012652 - 2025-06-28T17:50:03.012652 - 3.12.0 (tags/v3.12.0:0fb18b0, Oct 2 2023, 13:03:39) [MSC v.1935 64 bit (AMD64)]2025-06-28T17:50:03.012652 -
2025-06-28T17:50:03.012652 - ** Python executable:2025-06-28T17:50:03.012652 - 2025-06-28T17:50:03.012652 - E:\workspace\ComfyUI\venv312\Scripts\python.exe2025-06-28T17:50:03.012652 -
2025-06-28T17:50:03.012652 - ** ComfyUI Path:2025-06-28T17:50:03.012652 - 2025-06-28T17:50:03.012652 - E:\workspace\ComfyUI2025-06-28T17:50:03.012652 -
2025-06-28T17:50:03.012652 - ** ComfyUI Base Folder Path:2025-06-28T17:50:03.012652 - 2025-06-28T17:50:03.012652 - E:\workspace\ComfyUI2025-06-28T17:50:03.012652 -
2025-06-28T17:50:03.012652 - ** User directory:2025-06-28T17:50:03.012652 - 2025-06-28T17:50:03.012652 - E:\workspace\ComfyUI\user2025-06-28T17:50:03.012652 -
2025-06-28T17:50:03.012652 - ** ComfyUI-Manager config path:2025-06-28T17:50:03.012652 - 2025-06-28T17:50:03.012652 - E:\workspace\ComfyUI\user\default\ComfyUI-Manager\config.ini2025-06-28T17:50:03.012652 -
2025-06-28T17:50:03.012652 - ** Log path:2025-06-28T17:50:03.012652 - 2025-06-28T17:50:03.012652 - E:\workspace\ComfyUI\user\comfyui.log2025-06-28T17:50:03.012652 -
2025-06-28T17:50:03.870064 -
Prestartup times for custom nodes:
2025-06-28T17:50:03.870991 - 2.6 seconds: E:\workspace\ComfyUI\custom_nodes\comfyui-manager
2025-06-28T17:50:03.870991 -
2025-06-28T17:50:07.255486 - Checkpoint files will always be loaded safely.
2025-06-28T17:50:07.444962 - Total VRAM 6144 MB, total RAM 40661 MB
2025-06-28T17:50:07.445956 - pytorch version: 2.7.1+cu128
2025-06-28T17:50:07.446972 - Set vram state to: NORMAL_VRAM
2025-06-28T17:50:07.446972 - Device: cuda:0 NVIDIA GeForce RTX 3060 Laptop GPU : cudaMallocAsync
2025-06-28T17:50:09.677661 - Using pytorch attention
2025-06-28T17:50:12.437108 - Python version: 3.12.0 (tags/v3.12.0:0fb18b0, Oct 2 2023, 13:03:39) [MSC v.1935 64 bit (AMD64)]
2025-06-28T17:50:12.437108 - ComfyUI version: 0.3.41
2025-06-28T17:50:12.467140 - ComfyUI frontend version: 1.22.2
2025-06-28T17:50:12.469021 - [Prompt Server] web root: E:\workspace\ComfyUI\venv312\Lib\site-packages\comfyui_frontend_package\static
2025-06-28T17:50:14.832646 - ### Loading: ComfyUI-Manager (V3.33.1)
2025-06-28T17:50:14.836151 - [ComfyUI-Manager] network_mode: public
2025-06-28T17:50:15.045695 - ### ComfyUI Version: v0.3.41-11-g78f79266 | Released on '2025-06-22'
2025-06-28T17:50:15.388135 - [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/model-list.json
2025-06-28T17:50:15.392901 - [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/alter-list.json
2025-06-28T17:50:15.420693 - [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/github-stats.json
2025-06-28T17:50:15.480539 - [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/extension-node-map.json
2025-06-28T17:50:15.523083 - [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/custom-node-list.json
2025-06-28T17:50:15.547556 -
Import times for custom nodes:
2025-06-28T17:50:15.547556 - 0.0 seconds: E:\workspace\ComfyUI\custom_nodes\websocket_image_save.py
2025-06-28T17:50:15.547556 - 0.1 seconds: E:\workspace\ComfyUI\custom_nodes\comfyui-kjnodes
2025-06-28T17:50:15.548555 - 0.2 seconds: E:\workspace\ComfyUI\custom_nodes\comfyui-videohelpersuite
2025-06-28T17:50:15.548555 - 0.5 seconds: E:\workspace\ComfyUI\custom_nodes\comfyui-manager
2025-06-28T17:50:15.548555 - 1.1 seconds: E:\workspace\ComfyUI\custom_nodes\ComfyUI-LTXVideo
2025-06-28T17:50:15.548555 -
2025-06-28T17:50:16.213298 - Context impl SQLiteImpl.
2025-06-28T17:50:16.213298 - Will assume non-transactional DDL.
2025-06-28T17:50:16.217957 - No target revision found.
2025-06-28T17:50:16.231345 - Starting server
2025-06-28T17:50:16.232066 - To see the GUI go to: http://127.0.0.1:8188
2025-06-28T17:50:19.351277 - FETCH ComfyRegistry Data: 5/902025-06-28T17:50:19.351277 -
2025-06-28T17:50:22.550537 - got prompt
2025-06-28T17:50:22.554595 - Failed to validate prompt for output 1599:
2025-06-28T17:50:22.554595 - * LTXVLatentUpsamplerModelLoader 1828:
2025-06-28T17:50:22.554595 - - Value not in list: upscale_model: 'ltxv-spatial-upscaler-0.9.7.safetensors' not in []
2025-06-28T17:50:22.554595 - Output will be ignored
2025-06-28T17:50:22.966698 - model weight dtype torch.float8_e4m3fn, manual cast: torch.bfloat16
2025-06-28T17:50:22.972270 - model_type FLUX
2025-06-28T17:50:23.353566 - FETCH ComfyRegistry Data: 10/902025-06-28T17:50:23.353566 -
2025-06-28T17:50:41.242488 - FETCH ComfyRegistry Data: 15/902025-06-28T17:50:41.243488 -
2025-06-28T17:50:45.221827 - FETCH ComfyRegistry Data: 20/902025-06-28T17:50:45.222823 -
2025-06-28T17:50:47.571666 - VAE load device: cuda:0, offload device: cpu, dtype: torch.bfloat16
2025-06-28T17:50:47.571666 - no CLIP/text encoder weights in checkpoint, the text encoder model will not be loaded.
2025-06-28T17:50:49.441109 - FETCH ComfyRegistry Data: 25/902025-06-28T17:50:49.441109 -
2025-06-28T17:50:50.482613 - !!! Exception during processing !!! cannot access local variable 'self_attn_func' where it is not associated with a value
2025-06-28T17:50:50.483500 - Traceback (most recent call last):
File "E:\workspace\ComfyUI\execution.py", line 361, in execute
output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\workspace\ComfyUI\execution.py", line 236, in get_output_data
return_values = _map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\workspace\ComfyUI\execution.py", line 208, in _map_node_over_list
process_inputs(input_dict, i)
File "E:\workspace\ComfyUI\execution.py", line 197, in process_inputs
results.append(getattr(obj, func)(**inputs))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\workspace\ComfyUI\custom_nodes\ComfyUI-LTXVideo\q8_nodes.py", line 62, in patch
patcher(transformer, use_fp8_attention, True)
File "E:\workspace\ComfyUI\venv312\Lib\site-packages\q8_kernels\integration\patch_transformer.py", line 26, in patch_comfyui_native_transformer
attn_forward = cross_attn_forward(use_fp8_attention)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\workspace\ComfyUI\venv312\Lib\site-packages\q8_kernels\integration\comfyui_native.py", line 66, in cross_attn_forward
self_attn, cross_attn, is_out_tuple = get_attention_func(use_fp8_attention)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\workspace\ComfyUI\venv312\Lib\site-packages\q8_kernels\integration\utils.py", line 104, in get_attention_func
(self_attn_func, self_attn_memory_layout),
^^^^^^^^^^^^^^
UnboundLocalError: cannot access local variable 'self_attn_func' where it is not associated with a value
2025-06-28T17:50:50.490424 - Prompt executed in 27.93 seconds
2025-06-28T17:50:53.981873 - FETCH ComfyRegistry Data: 30/902025-06-28T17:50:53.981873 -
2025-06-28T17:51:01.741481 - FETCH ComfyRegistry Data: 35/902025-06-28T17:51:01.741481 -
2025-06-28T17:51:05.538151 - FETCH ComfyRegistry Data: 40/902025-06-28T17:51:05.538151 -
2025-06-28T17:51:09.085383 - got prompt
2025-06-28T17:51:09.090509 - Failed to validate prompt for output 1599:
2025-06-28T17:51:09.090509 - * LTXVLatentUpsamplerModelLoader 1828:
2025-06-28T17:51:09.090509 - - Value not in list: upscale_model: 'ltxv-spatial-upscaler-0.9.7.safetensors' not in []
2025-06-28T17:51:09.090509 - Output will be ignored
2025-06-28T17:51:09.105726 - !!! Exception during processing !!! LTXV Q8 Patcher is not applied to the model. Please use LTXQ8Patch node before loading lora or install q8_kernels.
2025-06-28T17:51:09.106709 - Traceback (most recent call last):
File "E:\workspace\ComfyUI\execution.py", line 361, in execute
output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\workspace\ComfyUI\execution.py", line 236, in get_output_data
return_values = _map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\workspace\ComfyUI\execution.py", line 208, in _map_node_over_list
process_inputs(input_dict, i)
File "E:\workspace\ComfyUI\execution.py", line 197, in process_inputs
results.append(getattr(obj, func)(**inputs))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\workspace\ComfyUI\custom_nodes\ComfyUI-LTXVideo\q8_nodes.py", line 135, in load_lora_model_only
return (self.load_lora(model, lora_name, strength_model),)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\workspace\ComfyUI\custom_nodes\ComfyUI-LTXVideo\q8_nodes.py", line 99, in load_lora
raise ValueError(
ValueError: LTXV Q8 Patcher is not applied to the model. Please use LTXQ8Patch node before loading lora or install q8_kernels.
2025-06-28T17:51:09.107717 - Prompt executed in 0.01 seconds
2025-06-28T17:51:10.931418 - FETCH ComfyRegistry Data: 45/902025-06-28T17:51:10.931418 -
2025-06-28T17:51:14.971682 - FETCH ComfyRegistry Data: 50/902025-06-28T17:51:14.972749 -
2025-06-28T17:51:46.809446 - FETCH ComfyRegistry Data: 55/902025-06-28T17:51:46.809446 -
2025-06-28T17:52:07.621921 - got prompt
2025-06-28T17:52:07.621921 - Failed to validate prompt for output 1599:
2025-06-28T17:52:07.621921 - * LTXVLatentUpsamplerModelLoader 1828:
2025-06-28T17:52:07.621921 - - Value not in list: upscale_model: 'ltxv-spatial-upscaler-0.9.7.safetensors' not in []
2025-06-28T17:52:07.621921 - Output will be ignored
2025-06-28T17:52:07.635691 - !!! Exception during processing !!! cannot access local variable 'self_attn_func' where it is not associated with a value
2025-06-28T17:52:07.643128 - Traceback (most recent call last):
File "E:\workspace\ComfyUI\execution.py", line 361, in execute
output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\workspace\ComfyUI\execution.py", line 236, in get_output_data
return_values = _map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\workspace\ComfyUI\execution.py", line 208, in _map_node_over_list
process_inputs(input_dict, i)
File "E:\workspace\ComfyUI\execution.py", line 197, in process_inputs
results.append(getattr(obj, func)(**inputs))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\workspace\ComfyUI\custom_nodes\ComfyUI-LTXVideo\q8_nodes.py", line 62, in patch
patcher(transformer, use_fp8_attention, True)
File "E:\workspace\ComfyUI\venv312\Lib\site-packages\q8_kernels\integration\patch_transformer.py", line 26, in patch_comfyui_native_transformer
attn_forward = cross_attn_forward(use_fp8_attention)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\workspace\ComfyUI\venv312\Lib\site-packages\q8_kernels\integration\comfyui_native.py", line 66, in cross_attn_forward
self_attn, cross_attn, is_out_tuple = get_attention_func(use_fp8_attention)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\workspace\ComfyUI\venv312\Lib\site-packages\q8_kernels\integration\utils.py", line 104, in get_attention_func
(self_attn_func, self_attn_memory_layout),
^^^^^^^^^^^^^^
UnboundLocalError: cannot access local variable 'self_attn_func' where it is not associated with a value
2025-06-28T17:52:07.644191 - Prompt executed in 0.01 seconds
Attached Workflow
Please make sure that workflow does not contain any sensitive information such as API keys or passwords.
Workflow too large. Please manually upload the workflow from local file system.
Additional Context
(Please add any additional context or steps to reproduce the error here)