AIMDO should work with multiple GPU's and the multiGPU node suite.
File "/home/gurila/Apps/ComfyUI/execution.py", line 535, 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/gurila/Apps/ComfyUI/execution.py", line 335, 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/gurila/Apps/ComfyUI/custom_nodes/comfyui-lora-manager/py/metadata_collector/metadata_hook.py", line 168, in async_map_node_over_list_with_metadata
results = await original_map_node_over_list(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
...<2 lines>...
)
^
File "/home/gurila/Apps/ComfyUI/execution.py", line 309, in _async_map_node_over_list
await process_inputs(input_dict, i)
File "/home/gurila/Apps/ComfyUI/execution.py", line 297, in process_inputs
result = f(**inputs)
File "/home/gurila/Apps/ComfyUI/nodes.py", line 80, in encode
return (clip.encode_from_tokens_scheduled(tokens), )
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^
File "/home/gurila/Apps/ComfyUI/comfy/sd.py", line 316, in encode_from_tokens_scheduled
pooled_dict = self.encode_from_tokens(tokens, return_pooled=return_pooled, return_dict=True)
File "/home/gurila/Apps/ComfyUI/comfy/sd.py", line 378, in encode_from_tokens
self.load_model(tokens)
~~~~~~~~~~~~~~~^^^^^^^^
File "/home/gurila/Apps/ComfyUI/comfy/sd.py", line 425, in load_model
model_management.load_models_gpu([self.patcher], memory_required=memory_used)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/gurila/Apps/ComfyUI/custom_nodes/ComfyUI-MultiGPU/distorch_2.py", line 211, in patched_load_models_gpu
loaded_model.model_load(lowvram_model_memory, force_patch_weights=force_patch_weights)
~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/gurila/Apps/ComfyUI/comfy/model_management.py", line 579, in model_load
self.model_use_more_vram(use_more_vram, force_patch_weights=force_patch_weights)
~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/gurila/Apps/ComfyUI/comfy/model_management.py", line 607, in model_use_more_vram
return self.model.partially_load(self.device, extra_memory, force_patch_weights=force_patch_weights)
~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/gurila/Apps/ComfyUI/comfy/model_patcher.py", line 1735, in partially_load
raise e
File "/home/gurila/Apps/ComfyUI/comfy/model_patcher.py", line 1732, in partially_load
self.load(device_to, dirty=dirty)
~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/gurila/Apps/ComfyUI/comfy/model_patcher.py", line 1558, in load
vbar = self._vbar_get(create=True)
File "/home/gurila/Apps/ComfyUI/comfy/model_patcher.py", line 1512, in _vbar_get
vbar = comfy_aimdo.model_vbar.ModelVBAR(self.model_size() * 10, self.load_device.index)
File "/home/gurila/Apps/ComfyUI/venv/lib/python3.14/site-packages/comfy_aimdo/model_vbar.py", line 49, in __init__
self._devctx = control.get_devctx(device)
~~~~~~~~~~~~~~~~~~^^^^^^^^
File "/home/gurila/Apps/ComfyUI/venv/lib/python3.14/site-packages/comfy_aimdo/control.py", line 113, in get_devctx
raise RuntimeError(f"comfy-aimdo device {device_id} is not initialized")
The MultiGPU node is trying to load the CLIP model onto cuda:1, but AIMDO was only
initialized on cuda:0.
The comfy_aimdo library does support multiple devices — see control.py:84 init_devices(device_ids). But ComfyUI's
main.py:221 only calls the single-device variant:
elif comfy_aimdo.control.init_device(comfy.model_management.get_torch_device().index):
That initializes AIMDO for cuda:0 only. When the MultiGPU node loads CLIP onto cuda:1, the new ModelPatcherDynamic calls
_vbar_get → get_devctx(1) → which raises because device 1 was never registered.
Option 1 — Disable AIMDO (quickest, no patching). Pick one when launching:
python main.py --disable-dynamic-vram
Option 2 — Patch main.py to init both GPUs (best of both worlds). Change line 221 from init_device(...) to
init_devices([0, 1]). AIMDO supports it; ComfyUI just doesn't call it that way. Caveat: changes a tracked file, so a git
pull may conflict.
Option 3 — Keep everything on cuda:0 in the workflow (move the CLIP loader's device back to cuda:0). Works but defeats
your multi-GPU setup.
Option 4 — File this upstream. This is arguably a ComfyUI bug — when MultiGPU is in play, AIMDO should be initialized for
every CUDA device, not just the primary.
RuntimeError: comfy-aimdo device 1 is not initialized
# ComfyUI Error Report
## Error Details
- **Node ID:** 4
- **Node Type:** CLIPTextEncode
- **Exception Type:** RuntimeError
- **Exception Message:** RuntimeError: comfy-aimdo device 1 is not initialized
## Stack Trace
File "/home/gurila/Apps/ComfyUI/execution.py", line 535, 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/gurila/Apps/ComfyUI/execution.py", line 335, 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/gurila/Apps/ComfyUI/custom_nodes/comfyui-lora-manager/py/metadata_collector/metadata_hook.py", line 168, in async_map_node_over_list_with_metadata
results = await original_map_node_over_list(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
...<2 lines>...
)
^
File "/home/gurila/Apps/ComfyUI/execution.py", line 309, in _async_map_node_over_list
await process_inputs(input_dict, i)
File "/home/gurila/Apps/ComfyUI/execution.py", line 297, in process_inputs
result = f(**inputs)
File "/home/gurila/Apps/ComfyUI/nodes.py", line 80, in encode
return (clip.encode_from_tokens_scheduled(tokens), )
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^
File "/home/gurila/Apps/ComfyUI/comfy/sd.py", line 316, in encode_from_tokens_scheduled
pooled_dict = self.encode_from_tokens(tokens, return_pooled=return_pooled, return_dict=True)
File "/home/gurila/Apps/ComfyUI/comfy/sd.py", line 378, in encode_from_tokens
self.load_model(tokens)
~~~~~~~~~~~~~~~^^^^^^^^
File "/home/gurila/Apps/ComfyUI/comfy/sd.py", line 425, in load_model
model_management.load_models_gpu([self.patcher], memory_required=memory_used)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/gurila/Apps/ComfyUI/custom_nodes/ComfyUI-MultiGPU/distorch_2.py", line 211, in patched_load_models_gpu
loaded_model.model_load(lowvram_model_memory, force_patch_weights=force_patch_weights)
~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/gurila/Apps/ComfyUI/comfy/model_management.py", line 579, in model_load
self.model_use_more_vram(use_more_vram, force_patch_weights=force_patch_weights)
~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/gurila/Apps/ComfyUI/comfy/model_management.py", line 607, in model_use_more_vram
return self.model.partially_load(self.device, extra_memory, force_patch_weights=force_patch_weights)
~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/gurila/Apps/ComfyUI/comfy/model_patcher.py", line 1735, in partially_load
raise e
File "/home/gurila/Apps/ComfyUI/comfy/model_patcher.py", line 1732, in partially_load
self.load(device_to, dirty=dirty)
~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/gurila/Apps/ComfyUI/comfy/model_patcher.py", line 1558, in load
vbar = self._vbar_get(create=True)
File "/home/gurila/Apps/ComfyUI/comfy/model_patcher.py", line 1512, in _vbar_get
vbar = comfy_aimdo.model_vbar.ModelVBAR(self.model_size() * 10, self.load_device.index)
File "/home/gurila/Apps/ComfyUI/venv/lib/python3.14/site-packages/comfy_aimdo/model_vbar.py", line 49, in __init__
self._devctx = control.get_devctx(device)
~~~~~~~~~~~~~~~~~~^^^^^^^^
File "/home/gurila/Apps/ComfyUI/venv/lib/python3.14/site-packages/comfy_aimdo/control.py", line 113, in get_devctx
raise RuntimeError(f"comfy-aimdo device {device_id} is not initialized")
## System Information
- **ComfyUI Version:** 0.20.1
- **Arguments:** main.py --listen 0.0.0.0 --enable-manager --use-sage-attention
- **OS:** linux
- **Python Version:** 3.14.4 (main, Apr 11 2026, 09:31:02) [GCC 15.2.1 20260209]
- **Embedded Python:** false
- **PyTorch Version:** 2.11.0+cu130
## Devices
- **Name:** cuda:0 NVIDIA GeForce RTX 5070 Ti : cudaMallocAsync
- **Type:** cuda
- **VRAM Total:** 16587751424
- **VRAM Free:** 15258615808
- **Torch VRAM Total:** 33554432
- **Torch VRAM Free:** 25034752
## Logs
2026-05-07T18:04:49.688286 - 2026-05-07T18:04:49.688311 - /home/gurila/Apps/ComfyUI/user/__manager/config.ini2026-05-07T18:04:49.688336 -
2026-05-07T18:04:49.688374 - ** Log path:2026-05-07T18:04:49.688398 - 2026-05-07T18:04:49.688423 - /home/gurila/Apps/ComfyUI/user/comfyui.log2026-05-07T18:04:49.688456 -
2026-05-07T18:04:49.869442 - [PRE] ComfyUI-Manager
2026-05-07T18:04:49.874843 -
Prestartup times for custom nodes:
2026-05-07T18:04:49.875071 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/comfyui-easy-use
2026-05-07T18:04:49.875163 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/ComfyUI-DepthAnythingV3
2026-05-07T18:04:49.875229 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/rgthree-comfy
2026-05-07T18:04:49.875291 -
2026-05-07T18:04:53.876850 - Found comfy_kitchen backend cuda: {'available': True, 'disabled': False, 'unavailable_reason': None, 'capabilities': ['apply_rope', 'apply_rope1', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'scaled_mm_nvfp4']}
2026-05-07T18:04:53.877096 - Found comfy_kitchen backend triton: {'available': True, 'disabled': True, 'unavailable_reason': None, 'capabilities': ['apply_rope', 'apply_rope1', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8']}
2026-05-07T18:04:53.877208 - Found comfy_kitchen backend eager: {'available': True, 'disabled': False, 'unavailable_reason': None, 'capabilities': ['apply_rope', 'apply_rope1', 'dequantize_mxfp8', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'scaled_mm_mxfp8', 'scaled_mm_nvfp4']}
2026-05-07T18:04:53.888069 - Checkpoint files will always be loaded safely.
2026-05-07T18:04:54.197958 - Total VRAM 15819 MB, total RAM 31924 MB
2026-05-07T18:04:54.198177 - pytorch version: 2.11.0+cu130
2026-05-07T18:04:54.198722 - Set vram state to: NORMAL_VRAM
2026-05-07T18:04:54.199201 - Device: cuda:0 NVIDIA GeForce RTX 5070 Ti : cudaMallocAsync
2026-05-07T18:04:54.199449 - Using async weight offloading with 2 streams
2026-05-07T18:04:54.199758 - Enabled pinned memory 28731.0
2026-05-07T18:04:54.717443 - Using sage attention
2026-05-07T18:04:57.698462 - DynamicVRAM support detected and enabled
2026-05-07T18:04:57.698680 - Python version: 3.14.4 (main, Apr 11 2026, 09:31:02) [GCC 15.2.1 20260209]
2026-05-07T18:04:57.698773 - ComfyUI version: 0.20.1
2026-05-07T18:04:57.700404 - comfy-aimdo version: 0.3.0
2026-05-07T18:04:57.702321 - comfy-kitchen version: 0.2.8
2026-05-07T18:04:57.705297 - ComfyUI frontend version: 1.43.17
2026-05-07T18:04:57.707430 - [Prompt Server] web root: /home/gurila/Apps/ComfyUI/venv/lib/python3.14/site-packages/comfyui_frontend_package/static
2026-05-07T18:04:57.711748 - Asset seeder disabled
2026-05-07T18:04:57.712259 - [START] ComfyUI-Manager
2026-05-07T18:04:57.794717 - [ComfyUI-Manager] Using GitPython backend2026-05-07T18:04:57.794790 -
2026-05-07T18:04:57.898118 - [ComfyUI-Manager] network_mode: public
2026-05-07T18:05:01.173250 - Blocked by policy: /home/gurila/Apps/ComfyUI/custom_nodes/comfyui-manager
2026-05-07T18:05:01.235770 - ### Loading: ComfyUI-Impact-Pack (V8.28.2)
2026-05-07T18:05:01.535879 - [Impact Pack] Wildcard total size (0.00 MB) is within cache limit (50.00 MB). Using full cache mode.
2026-05-07T18:05:01.537830 - [Impact Pack] Wildcards loading done.
2026-05-07T18:05:01.569600 -
2026-05-07T18:05:01.569704 - �[92m[rgthree-comfy] Loaded 48 extraordinary nodes. 🎉�[0m2026-05-07T18:05:01.569748 -
2026-05-07T18:05:01.569780 -
2026-05-07T18:05:01.569855 - �[33m[rgthree-comfy] ComfyUI's new Node 2.0 rendering may be incompatible with some rgthree-comfy nodes and features, breaking some rendering as well as losing the ability to access a node's properties (a vital part of many nodes). It also appears to run MUCH more slowly spiking CPU usage and causing jankiness and unresponsiveness, especially with large workflows. Personally I am not planning to use the new Nodes 2.0 and, unfortunately, am not able to invest the time to investigate and overhaul rgthree-comfy where needed. If you have issues when Nodes 2.0 is enabled, I'd urge you to switch it off as well and join me in hoping ComfyUI is not planning to deprecate the existing, stable canvas rendering all together.
�[0m2026-05-07T18:05:01.569896 -
2026-05-07T18:05:01.619188 - /home/gurila/Apps/ComfyUI/venv/lib/python3.14/site-packages/rotary_embedding_torch/rotary_embedding_torch.py:35: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
@autocast(enabled = False)
2026-05-07T18:05:01.619994 - /home/gurila/Apps/ComfyUI/venv/lib/python3.14/site-packages/rotary_embedding_torch/rotary_embedding_torch.py:268: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
@autocast(enabled = False)
2026-05-07T18:05:01.983537 - Using sage attention
2026-05-07T18:05:02.181739 - (RES4LYF) Init2026-05-07T18:05:02.181782 -
2026-05-07T18:05:02.182020 - (RES4LYF) Importing beta samplers.2026-05-07T18:05:02.182050 -
2026-05-07T18:05:02.217146 - (RES4LYF) Importing legacy samplers.2026-05-07T18:05:02.217220 -
2026-05-07T18:05:02.222847 - ### Loading: ComfyUI-Impact-Subpack (V1.3.5)
2026-05-07T18:05:02.226748 - [Impact Pack/Subpack] Using folder_paths to determine whitelist path: /home/gurila/Apps/ComfyUI/user/default/ComfyUI-Impact-Subpack/model-whitelist.txt
2026-05-07T18:05:02.227002 - [Impact Pack/Subpack] Ensured whitelist directory exists: /home/gurila/Apps/ComfyUI/user/default/ComfyUI-Impact-Subpack
2026-05-07T18:05:02.227519 - [Impact Pack/Subpack] Loaded 0 model(s) from whitelist: /home/gurila/Apps/ComfyUI/user/default/ComfyUI-Impact-Subpack/model-whitelist.txt
2026-05-07T18:05:02.324756 - [Impact Subpack] ultralytics_bbox: /home/gurila/Apps/ComfyUI/models/ultralytics/bbox
2026-05-07T18:05:02.324817 - [Impact Subpack] ultralytics_segm: /home/gurila/Apps/ComfyUI/models/ultralytics/segm
2026-05-07T18:05:04.547411 - �[34mWAS Node Suite: �[0mOpenCV Python FFMPEG support is enabled�[0m2026-05-07T18:05:04.547439 -
2026-05-07T18:05:04.547472 - �[34mWAS Node Suite �[93mWarning: �[0m`ffmpeg_bin_path` is not set in `/home/gurila/Apps/ComfyUI/custom_nodes/was-ns/was_suite_config.json` config file. Will attempt to use system ffmpeg binaries if available.�[0m2026-05-07T18:05:04.547482 -
2026-05-07T18:05:04.881927 - �[34mWAS Node Suite: �[0mFinished.�[0m �[32mLoaded�[0m �[0m220�[0m �[32mnodes successfully.�[0m2026-05-07T18:05:04.881953 -
2026-05-07T18:05:04.882012 -
�[3m�[93m"Art is a harmony parallel with nature."�[0m�[3m - Paul Cézanne�[0m
2026-05-07T18:05:04.882122 -
2026-05-07T18:05:04.903055 - [ComfyUI-DepthAnythingV3] Initializing custom node...2026-05-07T18:05:04.903093 -
2026-05-07T18:05:04.979047 - [ComfyUI-DepthAnythingV3] [OK] Node classes imported successfully2026-05-07T18:05:04.979123 -
2026-05-07T18:05:04.979162 - [ComfyUI-DepthAnythingV3] [OK] Loaded successfully!2026-05-07T18:05:04.979193 -
2026-05-07T18:05:04.996071 - �[1;34m[Mango Suite]: 🥭🥭🥭 �[96m�[3mResistor raving!�[0m 🥭🥭🥭2026-05-07T18:05:04.996209 -
2026-05-07T18:05:04.996318 - �[1;34m[Mango Suite]:�[0m Activated �[96m5�[0m file nodes.2026-05-07T18:05:04.996407 -
2026-05-07T18:05:05.110153 - USDU batch patches applied successfully.
2026-05-07T18:05:05.110230 - USDU batch patches applied successfully.
2026-05-07T18:05:05.134877 - [2026-05-07T18:05:05.134984 - UmeAiRT-Toolkit2026-05-07T18:05:05.135048 - ] 2026-05-07T18:05:05.135095 - Logic Nodes:2026-05-07T18:05:05.135145 - Could not import FaceDetailer internals: No module named '/home/gurila/Apps/ComfyUI/custom_nodes/comfyui-umeairt-toolkit.modules.facedetailer_core'2026-05-07T18:05:05.135199 -
2026-05-07T18:05:05.140198 - [2026-05-07T18:05:05.140280 - UmeAiRT-Toolkit2026-05-07T18:05:05.140331 - ] 🧩 Loaded 47 nodes.2026-05-07T18:05:05.140385 -
2026-05-07T18:05:05.140476 - [2026-05-07T18:05:05.140530 - UmeAiRT-Toolkit2026-05-07T18:05:05.140571 - ] 2026-05-07T18:05:05.140607 - ⚡ Optimisation check:2026-05-07T18:05:05.140646 - SageAttention 2026-05-07T18:05:05.140686 - ✅2026-05-07T18:05:05.140724 - | Flash Attention 2026-05-07T18:05:05.140890 - ✅2026-05-07T18:05:05.140928 - | Triton 2026-05-07T18:05:05.141188 - ✅2026-05-07T18:05:05.141239 -
2026-05-07T18:05:05.141583 - [2026-05-07T18:05:05.141624 - UmeAiRT-Toolkit2026-05-07T18:05:05.141657 - ] 2026-05-07T18:05:05.141693 - 📊 Initial CUDA memory:2026-05-07T18:05:05.141728 - 14.38GB free / 15.45GB total2026-05-07T18:05:05.141762 -
2026-05-07T18:05:05.141815 - [2026-05-07T18:05:05.141851 - UmeAiRT-Toolkit2026-05-07T18:05:05.141885 - ] 2026-05-07T18:05:05.141920 - ✅ Initialization Complete.2026-05-07T18:05:05.141973 -
2026-05-07T18:05:05.360915 - EA_LMStudio: Fetched 1 models from http://10.0.0.13:1234
2026-05-07T18:05:08.631548 - [ComfyUI-Easy-Use] server: 2026-05-07T18:05:08.631608 - v1.3.6 2026-05-07T18:05:08.631640 - Loaded2026-05-07T18:05:08.631675 -
2026-05-07T18:05:08.631715 - [ComfyUI-Easy-Use] web root: 2026-05-07T18:05:08.631743 - /home/gurila/Apps/ComfyUI/custom_nodes/comfyui-easy-use/web_version/v2 2026-05-07T18:05:08.631768 - Loaded2026-05-07T18:05:08.631792 -
2026-05-07T18:05:08.639462 - ComfyUI-GGUF: Allowing full torch compile
2026-05-07T18:05:08.983476 - WanVideoWrapper WARNING: FantasyPortrait nodes not available: Metaclasses with custom tp_new are not supported.
2026-05-07T18:05:09.005933 - IAMCCS compat: enabled local Qwen Multi-Gen seed widget patch
2026-05-07T18:05:09.031827 - [IAMCCS] WARNING: IAMCCS_QWEN_VL_FLF could not load — [IAMCCS_QWEN_VL_FLF] Cannot find QwenVLBase. Make sure ComfyUI-QwenVL is installed under custom_nodes/ComfyUI-QwenVL.2026-05-07T18:05:09.031908 -
2026-05-07T18:05:09.031995 - [IAMCCS] Make sure ComfyUI-QwenVL is installed in custom_nodes/ComfyUI-QwenVL2026-05-07T18:05:09.032041 -
2026-05-07T18:05:09.048433 -
___ _ __ __ ____ ____ ____ ____ _
|_ _| / \ | \/ |/ ___/ ___|/ ___| | _ \ ___ ___| | _____
| | / _ \ | |\/| | | | | \___ \ | |_) / _ \ / __| |/ / __|
| | / ___ \| | | | |__| |___ ___) || __/ (_) | (__| <\__ \
|___/_/ \_\_| |_|\____\____||____/ |_| \___/ \___|_|\_\___/
2026-05-07T18:05:09.048627 - by IAMCCS (follow me on patreon.com/IAMCCS or carminecristalloscalzi.com)
2026-05-07T18:05:09.049201 - IAMCCS nodes loaded: 161
2026-05-07T18:05:09.049323 - - IAMCCS-SuperNodes AU+IMG2VID Exec Finalize, IAMCCS-SuperNodes AU+IMG2VID Exec Planner, IAMCCS-SuperNodes AU+IMG2VID Exec Render, IAMCCS-SuperNodes AU+IMG2VID Exec VAE, IAMCCS-SuperNodes Second Stage, IAMCCS_AUIMG2VID_AudioTimeline, IAMCCS_AUIMG2VID_Continuity, IAMCCS_AUIMG2VID_DiskExtension, IAMCCS_AUIMG2VID_Finalize, IAMCCS_AUIMG2VID_KeyframeTimeline
2026-05-07T18:05:09.049412 - - IAMCCS_AUIMG2VID_Planner, IAMCCS_AUIMG2VID_ProjectTimelinePlanner, IAMCCS_AUIMG2VID_ReanchorLatent, IAMCCS_AUIMG2VID_RefreshPolicy, IAMCCS_AUIMG2VID_RuntimeBridge, IAMCCS_ApplyLoRAHooksToConditioning, IAMCCS_ApplyScheduledWanLoRAFromConditioning, IAMCCS_AudioExtender, IAMCCS_AudioExtensionMath, IAMCCS_AudioTimelineAssembler
2026-05-07T18:05:09.049488 - - IAMCCS_AudioTimelineGate, IAMCCS_AutoLinkArguments, IAMCCS_AutoLinkConverter, IAMCCS_BuildScheduledWanModelBank, IAMCCS_FluxKleinMultiGen, IAMCCS_GGUF_accelerator, IAMCCS_GetAutoLink, IAMCCS_HWProbeRecommendations, IAMCCS_HardMemoryPurge, IAMCCS_HwSupporter
2026-05-07T18:05:09.049554 - - IAMCCS_HwSupporterAny, IAMCCS_ImageResizeBatchSafe, IAMCCS_IntValueMonitor, IAMCCS_LTX2_AudioPromptDirector, IAMCCS_LTX2_BlendLatentBridge, IAMCCS_LTX2_CinematicLineStacker, IAMCCS_LTX2_CinematicMultiGenPlanner, IAMCCS_LTX2_CinematicPromptComposer, IAMCCS_LTX2_CinematicPromptRelayAdapter, IAMCCS_LTX2_CinematicRefLatentControl
2026-05-07T18:05:09.049619 - - IAMCCS_LTX2_CinematicShotAudioSelector, IAMCCS_LTX2_CinematicShotLineBuilder, IAMCCS_LTX2_CinematicShotPlanner, IAMCCS_LTX2_CinematicV2VAssetSelector, IAMCCS_LTX2_CinematicV2VTimelineLineBuilder, IAMCCS_LTX2_CinematicV2VTimelinePlanner, IAMCCS_LTX2_ConditionNextLatentWithPrevOverlap, IAMCCS_LTX2_ContextLatent, IAMCCS_LTX2_ControlPreprocess, IAMCCS_LTX2_EnsureFrames8nPlus1
2026-05-07T18:05:09.049691 - - IAMCCS_LTX2_EnsureMinFrames, IAMCCS_LTX2_ExtendSampler, IAMCCS_LTX2_ExtensionModule, IAMCCS_LTX2_ExtensionModule_Disk, IAMCCS_LTX2_ExtensionModule_simple, IAMCCS_LTX2_FirstLastFramesController, IAMCCS_LTX2_FirstLastLatentControl, IAMCCS_LTX2_FirstLastLatentControl_Pro, IAMCCS_LTX2_FrameCountValidator, IAMCCS_LTX2_FrameRateSync
2026-05-07T18:05:09.049754 - - IAMCCS_LTX2_GetImageFromBatch, IAMCCS_LTX2_ImageBatchCropByPad, IAMCCS_LTX2_ImageBatchPadReflect, IAMCCS_LTX2_InitLatentSampler, IAMCCS_LTX2_LastFrameBridgeLoad, IAMCCS_LTX2_LastFrameBridgeSave, IAMCCS_LTX2_LoRAStack, IAMCCS_LTX2_LoRAStackModelIO, IAMCCS_LTX2_LoRAStackSegmented6, IAMCCS_LTX2_LoRAStackStaged
2026-05-07T18:05:09.049815 - - IAMCCS_LTX2_LoadLatentBridge, IAMCCS_LTX2_LongVideoWrapperPrep, IAMCCS_LTX2_LongVideoWrapperPrepDisk, IAMCCS_LTX2_LoopingSampler, IAMCCS_LTX2_MiddleFrames, IAMCCS_LTX2_ModelWithLoRA_Segmented6, IAMCCS_LTX2_OneShotLowRAMLooper, IAMCCS_LTX2_ReferenceImageSwitch, IAMCCS_LTX2_ReferenceStartFramesInjector, IAMCCS_LTX2_SaveLatentBridge
2026-05-07T18:05:09.049877 - - IAMCCS_LTX2_SegmentQueueLoop, IAMCCS_LTX2_TimeFrameCount, IAMCCS_LTX2_Validator, IAMCCS_LatentTailSlice, IAMCCS_LazyAnySwitch, IAMCCS_LoadImagesFromDirLite, IAMCCS_LoadResizeSegmentFromDir, IAMCCS_Ltx2HelperModules_AudioTimeline, IAMCCS_Ltx2HelperModules_Continuity, IAMCCS_Ltx2HelperModules_DiskExtension
2026-05-07T18:05:09.049959 - - IAMCCS_Ltx2HelperModules_Finalize, IAMCCS_Ltx2HelperModules_KeyframeTimeline, IAMCCS_Ltx2HelperModules_Planner, IAMCCS_Ltx2HelperModules_ProjectTimelinePlanner, IAMCCS_Ltx2HelperModules_ReanchorLatent, IAMCCS_Ltx2HelperModules_RefreshPolicy, IAMCCS_Ltx2HelperModules_RuntimeBridge, IAMCCS_ModelWithLoRA, IAMCCS_ModelWithLoRA_LTX2, IAMCCS_ModelWithLoRA_LTX2_Staged
2026-05-07T18:05:09.050025 - - IAMCCS_ModelWithLoRA_RuntimeBridge, IAMCCS_MotionBridgeLoad, IAMCCS_MotionBridgeSave, IAMCCS_MotionScale, IAMCCS_MotionScaleAdvanced, IAMCCS_MoveAhead, IAMCCS_MoveAheadEnforcer, IAMCCS_MultiSwitch, IAMCCS_MultilinePromptSplitter8, IAMCCS_ProjectTimelinePlanner
2026-05-07T18:05:09.050079 - - IAMCCS_QwenMultiGen, IAMCCS_SamplerAdvancedVersion1, IAMCCS_SegmentPlanFromPlanner, IAMCCS_SegmentPlanner, IAMCCS_SegmentPlannerLinked, IAMCCS_SegmentPlannerSettings, IAMCCS_SegmentSwitch, IAMCCS_SelectScheduledWanModelFromConditioning, IAMCCS_SelectScheduledWanModelPairFromConditioning, IAMCCS_SetAutoLink
2026-05-07T18:05:09.050144 - - IAMCCS_SourceFramesToDisk, IAMCCS_SourceRangeFromSegmentPlan, IAMCCS_StartDirToVideoLatent, IAMCCS_StartImagesToVideoLatent, IAMCCS_SupernodeBase, IAMCCS_SupernodeModule, IAMCCS_ThreeSegmentPlanner, IAMCCS_TwoSegmentPlanner, IAMCCS_VAEDecodeTiledSafe, IAMCCS_VAEDecodeToDisk
2026-05-07T18:05:09.050203 - - IAMCCS_VRAMCleanup, IAMCCS_VRAMFlushLatent, IAMCCS_VideoCombineFromDir, IAMCCS_WanContinuityGuide, IAMCCS_WanImageBatchFrameSelect, IAMCCS_WanImageMotion, IAMCCS_WanImageMotionInductive, IAMCCS_WanImageMotionPro, IAMCCS_WanImageMotionProLegacy, IAMCCS_WanImageMotionProPlus
2026-05-07T18:05:09.050261 - - IAMCCS_WanImageMotionProPlus_Simple, IAMCCS_WanImageMotionPro_AdaIN, IAMCCS_WanImageMotion_AdaIN, IAMCCS_WanIndexedPromptEncode, IAMCCS_WanLoRAHookSchedule, IAMCCS_WanLoRARuntimeBridge, IAMCCS_WanLoRASchedule, IAMCCS_WanLoRAStack, IAMCCS_WanLoRAStackModelIO, IAMCCS_WanLongPlanner
2026-05-07T18:05:09.050314 - - IAMCCS_WanPrevTailPrep, IAMCCS_WanPromptLoopInfo, IAMCCS_WanPromptPhasePlanner, IAMCCS_WanSVIToFLFBridgeProPlus, IAMCCS_WanSVIToFLFBridgeProPlus_Simple, IAMCCS_bus_group, WanImageMotionPro, WanImageMotionProLegacy, WanImageMotionProPlus, WanMotionProTrimmer
2026-05-07T18:05:09.050372 - - iamccs_ltx2_lora_stack
2026-05-07T18:05:09.055136 - [MultiGPU Core Patching] Patching mm.soft_empty_cache for Comprehensive Memory Management (VRAM + CPU + Store Pruning)
2026-05-07T18:05:09.057178 - [MultiGPU] Patched comfy_kitchen CUDA DLPack wrapper with device guard.
2026-05-07T18:05:09.057687 - [MultiGPU Core Patching] Patching mm.get_torch_device, mm.text_encoder_device, mm.unet_offload_device
2026-05-07T18:05:09.058505 - [MultiGPU DEBUG] Initial current_device: cuda:0
2026-05-07T18:05:09.059554 - [MultiGPU DEBUG] Initial current_text_encoder_device: cuda:0
2026-05-07T18:05:09.060400 - [MultiGPU DEBUG] Initial current_unet_offload_device: cpu
2026-05-07T18:05:09.066725 - [MultiGPU] Initiating custom_node Registration. . .
2026-05-07T18:05:09.067229 - -----------------------------------------------
2026-05-07T18:05:09.067636 - custom_node Found Nodes
2026-05-07T18:05:09.068013 - -----------------------------------------------
2026-05-07T18:05:09.069658 - ComfyUI-LTXVideo Y 1
2026-05-07T18:05:09.070094 - ComfyUI-Florence2 N 0
2026-05-07T18:05:09.070476 - ComfyUI_bitsandbytes_NF4 Y 1
2026-05-07T18:05:09.070844 - x-flux-comfyui N 0
2026-05-07T18:05:09.071221 - ComfyUI-MMAudio N 0
2026-05-07T18:05:09.071577 - ComfyUI-GGUF Y 18
2026-05-07T18:05:09.071960 - PuLID_ComfyUI N 0
2026-05-07T18:05:09.072316 - ComfyUI-WanVideoWrapper Y 20
2026-05-07T18:05:09.072674 - -----------------------------------------------
2026-05-07T18:05:09.073232 - [MultiGPU] Registration complete. Final mappings: CheckpointLoaderAdvancedMultiGPU, CheckpointLoaderAdvancedDisTorch2MultiGPU, UNetLoaderLP, UNETLoaderMultiGPU, VAELoaderMultiGPU, CLIPLoaderMultiGPU, DualCLIPLoaderMultiGPU, TripleCLIPLoaderMultiGPU, QuadrupleCLIPLoaderMultiGPU, CLIPVisionLoaderMultiGPU, CheckpointLoaderSimpleMultiGPU, ControlNetLoaderMultiGPU, DiffusersLoaderMultiGPU, DiffControlNetLoaderMultiGPU, UNETLoaderDisTorch2MultiGPU, VAELoaderDisTorch2MultiGPU, CLIPLoaderDisTorch2MultiGPU, DualCLIPLoaderDisTorch2MultiGPU, TripleCLIPLoaderDisTorch2MultiGPU, QuadrupleCLIPLoaderDisTorch2MultiGPU, CLIPVisionLoaderDisTorch2MultiGPU, CheckpointLoaderSimpleDisTorch2MultiGPU, ControlNetLoaderDisTorch2MultiGPU, DiffusersLoaderDisTorch2MultiGPU, DiffControlNetLoaderDisTorch2MultiGPU, LTXVLoaderMultiGPU, CheckpointLoaderNF4MultiGPU, UnetLoaderGGUFDisTorchMultiGPU, UnetLoaderGGUFAdvancedDisTorchMultiGPU, CLIPLoaderGGUFDisTorchMultiGPU, DualCLIPLoaderGGUFDisTorchMultiGPU, TripleCLIPLoaderGGUFDisTorchMultiGPU, QuadrupleCLIPLoaderGGUFDisTorchMultiGPU, UnetLoaderGGUFDisTorch2MultiGPU, UnetLoaderGGUFAdvancedDisTorch2MultiGPU, CLIPLoaderGGUFDisTorch2MultiGPU, DualCLIPLoaderGGUFDisTorch2MultiGPU, TripleCLIPLoaderGGUFDisTorch2MultiGPU, QuadrupleCLIPLoaderGGUFDisTorch2MultiGPU, UnetLoaderGGUFMultiGPU, UnetLoaderGGUFAdvancedMultiGPU, CLIPLoaderGGUFMultiGPU, DualCLIPLoaderGGUFMultiGPU, TripleCLIPLoaderGGUFMultiGPU, QuadrupleCLIPLoaderGGUFMultiGPU, LoadWanVideoT5TextEncoderMultiGPU, WanVideoTextEncodeMultiGPU, WanVideoTextEncodeCachedMultiGPU, WanVideoTextEncodeSingleMultiGPU, WanVideoVAELoaderMultiGPU, WanVideoTinyVAELoaderMultiGPU, WanVideoBlockSwapMultiGPU, WanVideoImageToVideoEncodeMultiGPU, WanVideoDecodeMultiGPU, WanVideoModelLoaderMultiGPU, WanVideoSamplerMultiGPU, WanVideoVACEEncodeMultiGPU, WanVideoEncodeMultiGPU, LoadWanVideoClipTextEncoderMultiGPU, WanVideoClipVisionEncodeMultiGPU, WanVideoControlnetLoaderMultiGPU, FantasyTalkingModelLoaderMultiGPU, Wav2VecModelLoaderMultiGPU, WanVideoUni3C_ControlnetLoaderMultiGPU, DownloadAndLoadWav2VecModelMultiGPU
2026-05-07T18:05:09.087999 - [LoRA-Manager] Added path mapping: /mnt/drivefive1tb/Models/Comfy/models/Lora -> /home/gurila/Apps/ComfyUI/models/Lora
2026-05-07T18:05:09.090524 - [LoRA-Manager] Added path mapping: /mnt/drivefive1tb/Models/Comfy/models/loras -> /home/gurila/Apps/ComfyUI/models/loras
2026-05-07T18:05:09.092814 - [LoRA-Manager] Found LoRA roots:
- /home/gurila/Apps/ComfyUI/models/Lora
- /home/gurila/Apps/ComfyUI/models/loras
- /mnt/drivefive1tb/Models/Lora3
- /mnt/drivefour1tb/Models/lora2
2026-05-07T18:05:09.096542 - [LoRA-Manager] Added path mapping: /mnt/drivefive1tb/Models/Comfy/models/checkpoints -> /home/gurila/Apps/ComfyUI/models/checkpoints
2026-05-07T18:05:09.098847 - [LoRA-Manager] Added path mapping: /mnt/drivefive1tb/Models/Comfy/models/diffusion_models -> /home/gurila/Apps/ComfyUI/models/diffusion_models
2026-05-07T18:05:09.101162 - [LoRA-Manager] Added path mapping: /mnt/drivefive1tb/Models/Comfy/models/unet -> /home/gurila/Apps/ComfyUI/models/unet
2026-05-07T18:05:09.103300 - [LoRA-Manager] Found checkpoint roots:
- /home/gurila/Apps/ComfyUI/models/checkpoints
- /home/gurila/Apps/ComfyUI/models/diffusion_models
- /home/gurila/Apps/ComfyUI/models/unet
- /home/gurila/Apps/ComfyUI/output/checkpoints
- /home/gurila/Apps/ComfyUI/output/diffusion_models
- /mnt/cashstor2/mmodels/checkpoints2
- /mnt/cashstor2/mmodels/diffusion_models2
- /mnt/cashstor2/mmodels/unet2
2026-05-07T18:05:09.104865 - [LoRA-Manager] Added path mapping: /mnt/drivefive1tb/Models/Comfy/models/embeddings -> /home/gurila/Apps/ComfyUI/models/embeddings
2026-05-07T18:05:09.106330 - [LoRA-Manager] Found embedding roots:
- /home/gurila/Apps/ComfyUI/models/embeddings
2026-05-07T18:05:09.107250 - [LoRA-Manager] Symlink cache loaded with 0 mappings
2026-05-07T18:05:09.108003 - [LoRA-Manager] Symlink mappings restored from cache in 0.88 ms
2026-05-07T18:05:09.126051 - [LoRA-Manager] Updated 'comfyui' library with current folder paths
2026-05-07T18:05:09.331556 - [LoRA-Manager] Detected async ComfyUI execution, installing async metadata hooks
2026-05-07T18:05:09.331762 - [LoRA-Manager] Metadata collection hooks installed for runtime values
2026-05-07T18:05:09.331886 - [LoRA-Manager] ComfyUI Metadata Collector initialized
2026-05-07T18:05:09.332460 - [LoRA-Manager] Example images path:
2026-05-07T18:05:09.332847 - [LoRA-Manager] Added static route for locales: /locales -> /home/gurila/Apps/ComfyUI/custom_nodes/comfyui-lora-manager/locales
2026-05-07T18:05:09.592270 - [LayerComposer] Server routes registered ✅2026-05-07T18:05:09.592366 -
2026-05-07T18:05:09.595689 - ⚡ SeedVR2 optimizations check: SageAttention ✅ | Flash Attention ✅ | Triton ✅2026-05-07T18:05:09.595856 -
2026-05-07T18:05:09.740459 - 📊 Initial CUDA memory: 14.27GB free / 15.45GB total2026-05-07T18:05:09.740489 -
2026-05-07T18:05:09.779948 -
Import times for custom nodes:
2026-05-07T18:05:09.780032 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/ComfyUI-Wanify
2026-05-07T18:05:09.780055 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/ComfyUI-WanResolutionSelector
2026-05-07T18:05:09.780071 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/ComfyUI-Image-Conveyor
2026-05-07T18:05:09.780094 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/websocket_image_save.py
2026-05-07T18:05:09.780107 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/comfyui-image-tool
2026-05-07T18:05:09.780119 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/ComfyUI-QuickResolutionSelector
2026-05-07T18:05:09.780131 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/comfyui-zld
2026-05-07T18:05:09.780143 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/Compose-Plugin-Comfyui
2026-05-07T18:05:09.780154 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/comfy-lora-loader-with-triggerdb
2026-05-07T18:05:09.780165 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/ComfyUI_JPS-Nodes
2026-05-07T18:05:09.780176 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/ComfyUI-WanVideoAutoResize
2026-05-07T18:05:09.780188 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/comfyui-erenodes
2026-05-07T18:05:09.780199 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/Comfyui-ZiT-Lora-loader
2026-05-07T18:05:09.780218 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/zimagelatent
2026-05-07T18:05:09.780229 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/ComfyUI-GGUF
2026-05-07T18:05:09.780239 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/comfyui-qwen35
2026-05-07T18:05:09.780250 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/Comfy-Canvas
2026-05-07T18:05:09.780260 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/comfyui-inpaint-cropandstitch
2026-05-07T18:05:09.780270 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/cg-use-everywhere
2026-05-07T18:05:09.780283 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/mikey_nodes
2026-05-07T18:05:09.780294 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/deno-custom-nodes
2026-05-07T18:05:09.780305 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/unwdef-nodes-comfyui
2026-05-07T18:05:09.780316 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/ComfyUI-VFI
2026-05-07T18:05:09.780326 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/comfyui-mango-random
2026-05-07T18:05:09.780338 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/ComfyUI-Flux2Klein-Enhancer
2026-05-07T18:05:09.780348 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/comfyui-custom-scripts
2026-05-07T18:05:09.780359 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/comfyui_essentials
2026-05-07T18:05:09.780370 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/ComfyUI-MelBandRoFormer
2026-05-07T18:05:09.780381 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/comfyui_ultimatesdupscale
2026-05-07T18:05:09.780392 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/ComfyUI-JNodes
2026-05-07T18:05:09.780403 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/comfyui-umeairt-toolkit
2026-05-07T18:05:09.780414 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/ComfyUI-MultiGPU
2026-05-07T18:05:09.780429 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/comfyui-frame-interpolation
2026-05-07T18:05:09.780441 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/rgthree-comfy
2026-05-07T18:05:09.780456 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/comfyui_ipadapter_plus
2026-05-07T18:05:09.780467 - 0.0 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/comfyui-kjnodes
2026-05-07T18:05:09.780479 - 0.1 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/IAMCCS-nodes
2026-05-07T18:05:09.780491 - 0.1 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/ComfyUI_Swwan
2026-05-07T18:05:09.780503 - 0.1 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/matanyone
2026-05-07T18:05:09.780516 - 0.1 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/ComfyUI-DepthAnythingV3
2026-05-07T18:05:09.780528 - 0.1 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/ComfyUI-LTXVideo
2026-05-07T18:05:09.780540 - 0.1 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/comfyui_nvidia_rtx_nodes
2026-05-07T18:05:09.780551 - 0.1 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/comfyui-impact-subpack
2026-05-07T18:05:09.780562 - 0.1 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/comfyui-videohelpersuite
2026-05-07T18:05:09.780573 - 0.2 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/seedvr2_videoupscaler
2026-05-07T18:05:09.780585 - 0.2 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/EA_LMStudio
2026-05-07T18:05:09.780597 - 0.3 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/comfyui-lora-manager
2026-05-07T18:05:09.780609 - 0.3 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/comfyui-impact-pack
2026-05-07T18:05:09.780620 - 0.4 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/ComfyUI-WanVideoWrapper
2026-05-07T18:05:09.780631 - 0.5 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/RES4LYF
2026-05-07T18:05:09.780642 - 0.9 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/comfyui-easy-use
2026-05-07T18:05:09.780653 - 2.3 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/ComfyUI_bitsandbytes_NF4
2026-05-07T18:05:09.780665 - 2.6 seconds: /home/gurila/Apps/ComfyUI/custom_nodes/was-ns
2026-05-07T18:05:09.780678 -
2026-05-07T18:05:09.782499 - Context impl SQLiteImpl.
2026-05-07T18:05:09.782536 - Will assume non-transactional DDL.
2026-05-07T18:05:09.819871 - Starting server
2026-05-07T18:05:09.821786 - To see the GUI go to: http://0.0.0.0:8188
2026-05-07T18:05:09.824435 - [LoRA-Manager] Recipe cache miss: performing full directory scan
2026-05-07T18:05:09.832621 - [LoRA-Manager] Recipe cache initialized in 0.01 seconds. Found 0 recipes
2026-05-07T18:05:09.837780 - [LoRA-Manager] FTS index validated, reusing existing index with 0 recipes
2026-05-07T18:05:10.012385 - [LoRA-Manager] Lora cache hydrated from persisted snapshot with 1713 models
2026-05-07T18:05:10.014031 - [LoRA-Manager] Invalid cache entry for '/home/gurila/Apps/ComfyUI/models/checkpoints/flux-2-klein-9b-Q8_0.gguf': Required field 'sha256' is empty
2026-05-07T18:05:10.014174 - [LoRA-Manager] Invalid cache entry for '/home/gurila/Apps/ComfyUI/models/checkpoints/illustriousGehenna_v40.safetensors': Required field 'sha256' is empty
2026-05-07T18:05:10.014277 - [LoRA-Manager] Invalid cache entry for '/home/gurila/Apps/ComfyUI/models/diffusion_models/flux-2-klein-9b-fp8.safetensors': Required field 'sha256' is empty
2026-05-07T18:05:10.014373 - [LoRA-Manager] Invalid cache entry for '/home/gurila/Apps/ComfyUI/models/diffusion_models/randommergetime_v20.safetensors': Required field 'sha256' is empty
2026-05-07T18:05:10.014454 - [LoRA-Manager] Invalid cache entry for '/home/gurila/Apps/ComfyUI/models/diffusion_models/zTurboCreartUltimate_zCreartUltimatev2.safetensors': Required field 'sha256' is empty
2026-05-07T18:05:10.014563 - [LoRA-Manager] Invalid cache entry for '/home/gurila/Apps/ComfyUI/models/diffusion_models/z_image_turbo_nvfp4.safetensors': Required field 'sha256' is empty
2026-05-07T18:05:10.014637 - [LoRA-Manager] Invalid cache entry for '/home/gurila/Apps/ComfyUI/models/unet/z-image-Q8_0.gguf': Required field 'sha256' is empty
2026-05-07T18:05:10.014877 - [LoRA-Manager] Cache health check: corrupted - 7/8 invalid, 0 repaired
2026-05-07T18:05:10.015035 - [LoRA-Manager] Broadcasting cache health warning: corrupted (7 invalid entries)
2026-05-07T18:05:10.015131 - [LoRA-Manager] Checkpoint Scanner: Cache health issue detected - 7 invalid entries, 0 repaired
2026-05-07T18:05:10.015433 - [LoRA-Manager] Checkpoint cache hydrated from persisted snapshot with 1 models
2026-05-07T18:05:10.064669 - [LoRA-Manager] Embedding cache hydrated from persisted snapshot with 480 models
2026-05-07T18:06:10.462454 - [DEPRECATION WARNING] Detected import of deprecated legacy API: /scripts/ui.js. This is likely caused by a custom node extension using outdated APIs. Please update your extensions or contact the extension author for an updated version.
2026-05-07T18:06:10.468823 - [DEPRECATION WARNING] Detected import of deprecated legacy API: /extensions/core/clipspace.js. This is likely caused by a custom node extension using outdated APIs. Please update your extensions or contact the extension author for an updated version.
2026-05-07T18:06:10.470799 - [ComfyUI-Manager] The ComfyRegistry cache update is still in progress, so an outdated cache is being used.
2026-05-07T18:06:10.859082 - FETCH DATA from: /home/gurila/Apps/ComfyUI/user/__manager/cache/3351870820_custom-node-list.json2026-05-07T18:06:10.881598 - [DONE]2026-05-07T18:06:10.881963 -
2026-05-07T18:06:10.973844 - [DEPRECATION WARNING] Detected import of deprecated legacy API: /extensions/core/widgetInputs.js. This is likely caused by a custom node extension using outdated APIs. Please update your extensions or contact the extension author for an updated version.
2026-05-07T18:06:13.896518 - ComfyUI-GGUF: Allowing full torch compile
2026-05-07T18:06:13.920022 - [UmeAiRT-Toolkit] SeedVR2: Using vendored copy2026-05-07T18:06:13.920181 -
2026-05-07T18:06:23.558517 - got prompt
2026-05-07T18:06:23.670827 - [MultiGPU Core Patching] Successfully patched ModelPatcher.partially_load
2026-05-07T18:06:23.676559 - Using pytorch attention in VAE
2026-05-07T18:06:23.677146 - Using pytorch attention in VAE
2026-05-07T18:06:23.717259 - VAE load device: cuda:0, offload device: cpu, dtype: torch.bfloat16
2026-05-07T18:06:23.718040 - [MultiGPU DisTorch V2] Full allocation string: #cuda:0;4.0;cuda:1
2026-05-07T18:06:23.739269 - [MultiGPU Core Patching] text_encoder_device_patched returning device: cuda:1 (current_text_encoder_device=cuda:1)
2026-05-07T18:06:23.869036 - CLIP/text encoder model load device: cuda:1, offload device: cpu, current: cpu, dtype: torch.float16
2026-05-07T18:06:23.883998 - Found quantization metadata version 1
2026-05-07T18:06:23.884076 - Detected mixed precision quantization
2026-05-07T18:06:23.884227 - Using mixed precision operations
2026-05-07T18:06:23.899807 - model weight dtype torch.float8_e4m3fn, manual cast: torch.bfloat16
2026-05-07T18:06:23.903652 - model_type FLUX
2026-05-07T18:06:24.126429 - Requested to load PixArtTEModel_
2026-05-07T18:06:24.167513 - !!! Exception during processing !!! comfy-aimdo device 1 is not initialized
2026-05-07T18:06:24.181108 - Traceback (most recent call last):
File "/home/gurila/Apps/ComfyUI/execution.py", line 535, 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/gurila/Apps/ComfyUI/execution.py", line 335, 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/gurila/Apps/ComfyUI/custom_nodes/comfyui-lora-manager/py/metadata_collector/metadata_hook.py", line 168, in async_map_node_over_list_with_metadata
results = await original_map_node_over_list(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
...<2 lines>...
)
^
File "/home/gurila/Apps/ComfyUI/execution.py", line 309, in _async_map_node_over_list
await process_inputs(input_dict, i)
File "/home/gurila/Apps/ComfyUI/execution.py", line 297, in process_inputs
result = f(**inputs)
File "/home/gurila/Apps/ComfyUI/nodes.py", line 80, in encode
return (clip.encode_from_tokens_scheduled(tokens), )
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^
File "/home/gurila/Apps/ComfyUI/comfy/sd.py", line 316, in encode_from_tokens_scheduled
pooled_dict = self.encode_from_tokens(tokens, return_pooled=return_pooled, return_dict=True)
File "/home/gurila/Apps/ComfyUI/comfy/sd.py", line 378, in encode_from_tokens
self.load_model(tokens)
~~~~~~~~~~~~~~~^^^^^^^^
File "/home/gurila/Apps/ComfyUI/comfy/sd.py", line 425, in load_model
model_management.load_models_gpu([self.patcher], memory_required=memory_used)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/gurila/Apps/ComfyUI/custom_nodes/ComfyUI-MultiGPU/distorch_2.py", line 211, in patched_load_models_gpu
loaded_model.model_load(lowvram_model_memory, force_patch_weights=force_patch_weights)
~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/gurila/Apps/ComfyUI/comfy/model_management.py", line 579, in model_load
self.model_use_more_vram(use_more_vram, force_patch_weights=force_patch_weights)
~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/gurila/Apps/ComfyUI/comfy/model_management.py", line 607, in model_use_more_vram
return self.model.partially_load(self.device, extra_memory, force_patch_weights=force_patch_weights)
~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/gurila/Apps/ComfyUI/comfy/model_patcher.py", line 1735, in partially_load
raise e
File "/home/gurila/Apps/ComfyUI/comfy/model_patcher.py", line 1732, in partially_load
self.load(device_to, dirty=dirty)
~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/gurila/Apps/ComfyUI/comfy/model_patcher.py", line 1558, in load
vbar = self._vbar_get(create=True)
File "/home/gurila/Apps/ComfyUI/comfy/model_patcher.py", line 1512, in _vbar_get
vbar = comfy_aimdo.model_vbar.ModelVBAR(self.model_size() * 10, self.load_device.index)
File "/home/gurila/Apps/ComfyUI/venv/lib/python3.14/site-packages/comfy_aimdo/model_vbar.py", line 49, in __init__
self._devctx = control.get_devctx(device)
~~~~~~~~~~~~~~~~~~^^^^^^^^
File "/home/gurila/Apps/ComfyUI/venv/lib/python3.14/site-packages/comfy_aimdo/control.py", line 113, in get_devctx
raise RuntimeError(f"comfy-aimdo device {device_id} is not initialized")
RuntimeError: comfy-aimdo device 1 is not initialized
2026-05-07T18:06:24.186872 - Prompt executed in 0.61 seconds
2026-05-07T18:06:52.623375 - FETCH ComfyRegistry Data [DONE]
2026-05-07T18:06:52.924447 - [ComfyUI-Manager] default cache updated: https://api.comfy.org/nodes
2026-05-07T18:06:52.993829 - FETCH DATA from: /home/gurila/Apps/ComfyUI/user/__manager/cache/3351870820_custom-node-list.json2026-05-07T18:06:53.011805 - [DONE]2026-05-07T18:06:53.011892 -
2026-05-07T18:06:53.055713 - [ComfyUI-Manager] All startup tasks have been completed.
## 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)
Custom Node Testing
Expected Behavior
AIMDO should work with multiple GPU's and the multiGPU node suite.
Actual Behavior
ComfyUI Error Report
Error Details
Stack Trace
System Information
Devices
Steps to Reproduce
The MultiGPU node is trying to load the CLIP model onto cuda:1, but AIMDO was only
initialized on cuda:0.
Root cause
The comfy_aimdo library does support multiple devices — see control.py:84 init_devices(device_ids). But ComfyUI's
main.py:221 only calls the single-device variant:
elif comfy_aimdo.control.init_device(comfy.model_management.get_torch_device().index):
That initializes AIMDO for cuda:0 only. When the MultiGPU node loads CLIP onto cuda:1, the new ModelPatcherDynamic calls
_vbar_get → get_devctx(1) → which raises because device 1 was never registered.
Option 1 — Disable AIMDO (quickest, no patching). Pick one when launching:
python main.py --disable-dynamic-vram
or any of these also bypass AIMDO per cli_args.py:272:
python main.py --highvram
python main.py --gpu-only
You lose DynamicVRAM but the legacy ModelPatcher works fine with MultiGPU.
Option 2 — Patch main.py to init both GPUs (best of both worlds). Change line 221 from init_device(...) to
init_devices([0, 1]). AIMDO supports it; ComfyUI just doesn't call it that way. Caveat: changes a tracked file, so a git
pull may conflict.
Option 3 — Keep everything on cuda:0 in the workflow (move the CLIP loader's device back to cuda:0). Works but defeats
your multi-GPU setup.
Option 4 — File this upstream. This is arguably a ComfyUI bug — when MultiGPU is in play, AIMDO should be initialized for
every CUDA device, not just the primary.
Debug Logs
Other
No response