GGUF插件加载QWEN模型生成黑图问题 #10800
Description
Custom Node Testing
- I have tried disabling custom nodes and the issue persists (see how to disable custom nodes if you need help)
Your question
GGUF插件加载QWEN GGUF模型生成黑图问题，日志如下：cd ./ComfyUI && PORT=8188 export PYTHONMALLOC=malloc
export MALLOC_TRIM_THRESHOLD_=100000
export MALLOC_ARENA_MAX=2
export PYTORCH_CUDA_ALLOC_CONF=max_split_size_mb:64,garbage_collection_threshold:0.4
export CUDA_VISIBLE_DEVICES=0
export OMP_NUM_THREADS=2
export MKL_NUM_THREADS=2
python main.py --listen 0.0.0.0 --port 8188 --normalvram --disable-auto-launch --disable-metadata --dont-upcast-attention --reserve-vram 1.0 --force-fp16 --fp16-vae
pyenv shell 3.11.1
➜ /workspace cd ./ComfyUI && PORT=8188 export PYTHONMALLOC=malloc
➜ ComfyUI export MALLOC_TRIM_THRESHOLD_=100000
➜ ComfyUI export MALLOC_ARENA_MAX=2
➜ ComfyUI export PYTORCH_CUDA_ALLOC_CONF=max_split_size_mb:64,garbage_collection_threshold:0.4
➜ ComfyUI export CUDA_VISIBLE_DEVICES=0
➜ ComfyUI export OMP_NUM_THREADS=2
➜ ComfyUI export MKL_NUM_THREADS=2
➜ ComfyUI python main.py --listen 0.0.0.0 --port 8188 --normalvram --disable-auto-launch --disable-metadata --dont-upcast-attention --r
eserve-vram 1.0 --force-fp16 --fp16-vae
[START] Security scan
[DONE] Security scan
ComfyUI-Manager: installing dependencies done.
** ComfyUI startup time: 2025-11-19 16:08:05.755
** Platform: Linux
** Python version: 3.11.1 (main, Sep 17 2025, 11:40:12) [GCC 13.3.0]
** Python executable: /root/.pyenv/versions/3.11.1/bin/python
** ComfyUI Path: /workspace/ComfyUI
** ComfyUI Base Folder Path: /workspace/ComfyUI
** User directory: /workspace/ComfyUI/user
** ComfyUI-Manager config path: /workspace/ComfyUI/user/default/ComfyUI-Manager/config.ini
** Log path: /workspace/ComfyUI/user/comfyui.log
Prestartup times for custom nodes:
0.0 seconds: /workspace/ComfyUI/custom_nodes/rgthree-comfy
0.0 seconds: /workspace/ComfyUI/custom_nodes/comfyui-easy-use
5.0 seconds: /workspace/ComfyUI/custom_nodes/ComfyUI-Manager
/root/.pyenv/versions/3.11.1/lib/python3.11/site-packages/torch/cuda/init.py:63: FutureWarning: The pynvml package is deprecated. Please install nvidia-ml-py instead. If you did not install pynvml directly, please report this to the maintainers of the package that installed pynvml for you.
import pynvml # type: ignore[import]
Checkpoint files will always be loaded safely.
Total VRAM 15931 MB, total RAM 31711 MB
pytorch version: 2.9.1+cu128
Set vram state to: NORMAL_VRAM
Device: cuda:0 Tesla T4 : cudaMallocAsync
Enabled pinned memory 30125.0
working around nvidia conv3d memory bug.
Using pytorch attention
Python version: 3.11.1 (main, Sep 17 2025, 11:40:12) [GCC 13.3.0]
ComfyUI version: 0.3.70
ComfyUI frontend version: 1.28.8
[Prompt Server] web root: /root/.pyenv/versions/3.11.1/lib/python3.11/site-packages/comfyui_frontend_package/static
Loading: ComfyUI-Manager (V3.37.1)
[ComfyUI-Manager] network_mode: offline
ComfyUI Version: v0.3.70-2-g65ee24c9 | Released on '2025-11-19'
[ComfyUI-Manager] All startup tasks have been completed.
Loading: ComfyUI-Impact-Pack (V8.28)
[Impact Pack] Wildcard total size (0.00 MB) is within cache limit (50.00 MB). Using full cache mode.
[Impact Pack] Wildcards loading done.
[/workspace/ComfyUI/custom_nodes/comfy-mtb] | INFO -> loaded 106 nodes successfuly
[/workspace/ComfyUI/custom_nodes/comfy-mtb] | INFO -> Some nodes (2) could not be loaded. This can be ignored, but go to http://None:8188/mtb if you want more information.
[ComfyUI-Easy-Use] server: v1.3.4 Loaded
[ComfyUI-Easy-Use] web root: /workspace/ComfyUI/custom_nodes/comfyui-easy-use/web_version/v2 Loaded
》===>====>========>Fk_Server:OK！<========<====<===《
FKServer: 资源上传服务器启动成功
FKServer: 接口APIurl：http://localhost:3059
FKServer: 上传目录: /workspace/ComfyUI/input
FKServer: python: /root/.pyenv/versions/3.11.1/bin/python
Warning: Could not load sageattention: No module named 'sageattention'
sageattention package is not installed, sageattention will not be available
[/workspace/ComfyUI/custom_nodes/comfyui_controlnet_aux] | INFO -> Using ckpts path: /workspace/ComfyUI/custom_nodes/comfyui_controlnet_aux/ckpts
[/workspace/ComfyUI/custom_nodes/comfyui_controlnet_aux] | INFO -> Using symlinks: False
[/workspace/ComfyUI/custom_nodes/comfyui_controlnet_aux] | INFO -> Using ort providers: ['CUDAExecutionProvider', 'DirectMLExecutionProvider', 'OpenVINOExecutionProvider', 'ROCMExecutionProvider', 'CPUExecutionProvider', 'CoreMLExecutionProvider']
DWPose: Onnxruntime with acceleration providers detected
ComfyUI-GGUF: Allowing full torch compile
Mixlab Nodes: Loaded
json_repair## OK
ChatGPT.available True
edit_mask.available True
clip_interrogator_model not found: /workspace/ComfyUI/models/clip_interrogator/Salesforce/blip-image-captioning-base, pls download from https://huggingface.co/Salesforce/blip-image-captioning-base
ClipInterrogator.available True
text_generator_model not found: /workspace/ComfyUI/models/prompt_generator/text2image-prompt-generator, pls download from https://huggingface.co/succinctly/text2image-prompt-generator/tree/main
zh_en_model not found: /workspace/ComfyUI/models/prompt_generator/opus-mt-zh-en, pls download from https://huggingface.co/Helsinki-NLP/opus-mt-zh-en/tree/main
PromptGenerate.available True
ChinesePrompt.available True
RembgNode_.available True
ffmpeg could not be found. Using ffmpeg from imageio-ffmpeg.
TripoSR.available
MiniCPMNode.available
Scenedetect.available
FishSpeech.available
SenseVoice.available
Whisper.available False
fal-client## OK
FalVideo.available
[rgthree-comfy] Loaded 48 exciting nodes. 🎉
[Crystools INFO] Crystools version: 1.27.4
[Crystools INFO] Platform release: 6.6.92-34.1.tl4.x86_64
[Crystools INFO] JETSON: Not detected.
[Crystools INFO] CPU: Intel(R) Xeon(R) Platinum 8255C CPU @ 2.50GHz - Arch: x86_64 - OS: Linux 6.6.92-34.1.tl4.x86_64
[Crystools INFO] pynvml (NVIDIA) initialized.
[Crystools INFO] GPU/s:
[Crystools INFO] 0) Tesla T4
[Crystools INFO] NVIDIA Driver: 580.65.06
/root/.pyenv/versions/3.11.1/lib/python3.11/site-packages/timm/models/layers/init.py:48: FutureWarning: Importing from timm.models.layers is deprecated, please import via timm.layers
warnings.warn(f"Importing from {name} is deprecated, please import via timm.layers", FutureWarning)
[ComfyUI-RMBG] v2.9.0 | 32 nodes Loaded
Comfyroll Studio v1.76 : 175 Nodes Loaded
** For changes, please see patch notes at https://github.com/Suzie1/ComfyUI_Comfyroll_CustomNodes/blob/main/Patch_Notes.md
** For help, please see the wiki at https://github.com/Suzie1/ComfyUI_Comfyroll_CustomNodes/wiki
2025-11-19 16:08:25.401 | INFO | /workspace/ComfyUI/custom_nodes/hunyuanvideo-foley::11 - Loading HunyuanVideo-Foley nodes with FP8 quantization and torch.compile support
2025-11-19 16:08:25.542 | ERROR | /workspace/ComfyUI/custom_nodes/hunyuanvideo-foley.nodes::58 - Failed to import HunyuanVideo-Foley modules: No module named 'decord'
2025-11-19 16:08:25.542 | ERROR | /workspace/ComfyUI/custom_nodes/hunyuanvideo-foley.nodes::59 - Make sure the HunyuanVideo-Foley package is installed and accessible
Traceback (most recent call last):
File "/workspace/ComfyUI/nodes.py", line 2131, in load_custom_node
module_spec.loader.exec_module(module)
File "", line 940, in exec_module
File "", line 241, in _call_with_frames_removed
File "/workspace/ComfyUI/custom_nodes/hunyuanvideo-foley/init.py", line 12, in
from .nodes import NODE_CLASS_MAPPINGS, NODE_DISPLAY_NAME_MAPPINGS
File "/workspace/ComfyUI/custom_nodes/hunyuanvideo-foley/nodes.py", line 53, in
from .utils import denoise_process_safely
File "/workspace/ComfyUI/custom_nodes/hunyuanvideo-foley/utils.py", line 11, in
import decord
ModuleNotFoundError: No module named 'decord'
Cannot import /workspace/ComfyUI/custom_nodes/hunyuanvideo-foley module for custom nodes: No module named 'decord'
✨提示词小助手 V1.2.3 已启动
/root/.pyenv/versions/3.11.1/lib/python3.11/site-packages/pydub/utils.py:170: RuntimeWarning: Couldn't find ffmpeg or avconv - defaulting to ffmpeg, but may not work
warn("Couldn't find ffmpeg or avconv - defaulting to ffmpeg, but may not work", RuntimeWarning)
Please 'pip install xformers'
Nvidia APEX normalization not installed, using PyTorch LayerNorm
Import times for custom nodes:
0.0 seconds: /workspace/ComfyUI/custom_nodes/websocket_image_save.py
0.0 seconds: /workspace/ComfyUI/custom_nodes/ComfyUI_AdvancedRefluxControl
0.0 seconds: /workspace/ComfyUI/custom_nodes/ComfyUI-DD-Translation
0.0 seconds: /workspace/ComfyUI/custom_nodes/ComfyUI-GGUF
0.0 seconds: /workspace/ComfyUI/custom_nodes/cg-use-everywhere
0.0 seconds: /workspace/ComfyUI/custom_nodes/wavespeed
0.0 seconds: /workspace/ComfyUI/custom_nodes/comfyui_essentials
0.0 seconds: /workspace/ComfyUI/custom_nodes/comfyui-custom-scripts
0.0 seconds: /workspace/ComfyUI/custom_nodes/prompt-assistant
0.0 seconds: /workspace/ComfyUI/custom_nodes/comfyui-Pollinations
0.0 seconds: /workspace/ComfyUI/custom_nodes/comfyui-videohelpersuite
0.0 seconds: /workspace/ComfyUI/custom_nodes/rgthree-comfy
0.0 seconds: /workspace/ComfyUI/custom_nodes/comfyui-kjnodes
0.0 seconds: /workspace/ComfyUI/custom_nodes/ComfyUI_Comfyroll_CustomNodes
0.0 seconds: /workspace/ComfyUI/custom_nodes/comfyui_controlnet_aux
0.1 seconds: /workspace/ComfyUI/custom_nodes/comfyui_layerstyle
0.1 seconds: /workspace/ComfyUI/custom_nodes/ComfyUI-Manager
0.1 seconds: /workspace/ComfyUI/custom_nodes/comfyui-rmbg
0.1 seconds (IMPORT FAILED): /workspace/ComfyUI/custom_nodes/hunyuanvideo-foley
0.1 seconds: /workspace/ComfyUI/custom_nodes/ComfyUI-Crystools
0.1 seconds: /workspace/ComfyUI/custom_nodes/ComfyUI-WanVideoWrapper
0.2 seconds: /workspace/ComfyUI/custom_nodes/comfyui_pulid_flux_ll
0.2 seconds: /workspace/ComfyUI/custom_nodes/comfyui_fk_server
0.2 seconds: /workspace/ComfyUI/custom_nodes/comfyui-impact-pack
0.4 seconds: /workspace/ComfyUI/custom_nodes/ComfyUI-ModelScope-API
1.8 seconds: /workspace/ComfyUI/custom_nodes/comfy-mtb
2.1 seconds: /workspace/ComfyUI/custom_nodes/comfyui-easy-use
5.1 seconds: /workspace/ComfyUI/custom_nodes/comfyui-mixlab-nodes
Context impl SQLiteImpl.
Will assume non-transactional DDL.
No target revision found.
Starting server
To see the GUI go to: http://0.0.0.0:8188
FKServer: ApiID: GNH0OQIUObGndx3epIKcnVdnuIaNxgT5
[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.
[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.
[DEPRECATION WARNING] Detected import of deprecated legacy API: /extensions/core/groupNode.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.
/workspace/ComfyUI/custom_nodes/comfyui-mixlab-nodes/webApp/lib/photoswipe-lightbox.esm.min.js
[DEPRECATION WARNING] Detected import of deprecated legacy API: /scripts/ui/components/buttonGroup.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.
[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.
/workspace/ComfyUI/custom_nodes/comfyui-mixlab-nodes/webApp/lib/pickr.min.js
/workspace/ComfyUI/custom_nodes/comfyui-mixlab-nodes/webApp/lib/photoswipe.min.css
/workspace/ComfyUI/custom_nodes/comfyui-mixlab-nodes/webApp/lib/juxtapose.min.js
/workspace/ComfyUI/custom_nodes/comfyui-mixlab-nodes/webApp/lib/model-viewer.min.js
/workspace/ComfyUI/custom_nodes/comfyui-mixlab-nodes/webApp/lib/classic.min.css
/workspace/ComfyUI/custom_nodes/comfyui-mixlab-nodes/webApp/lib/juxtapose.css
[DEPRECATION WARNING] Detected import of deprecated legacy API: /scripts/ui/components/button.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.
/workspace/ComfyUI/custom_nodes/comfyui-mixlab-nodes/webApp/lib/photoswipe-lightbox.esm.min.js
/workspace/ComfyUI/custom_nodes/comfyui-mixlab-nodes/webApp/lib/pickr.min.js
/workspace/ComfyUI/custom_nodes/comfyui-mixlab-nodes/webApp/lib/photoswipe.min.css
/workspace/ComfyUI/custom_nodes/comfyui-mixlab-nodes/webApp/lib/juxtapose.min.js
/workspace/ComfyUI/custom_nodes/comfyui-mixlab-nodes/webApp/lib/model-viewer.min.js
/workspace/ComfyUI/custom_nodes/comfyui-mixlab-nodes/webApp/lib/classic.min.css
/workspace/ComfyUI/custom_nodes/comfyui-mixlab-nodes/webApp/lib/juxtapose.css
got prompt
Using pytorch attention in VAE
Using pytorch attention in VAE
VAE load device: cuda:0, offload device: cpu, dtype: torch.float16
Using scaled fp8: fp8 matrix mult: False, scale input: False
CLIP/text encoder model load device: cuda:0, offload device: cpu, current: cpu, dtype: torch.float16
Requested to load QwenImageTEModel_
loaded completely; 13982.42 MB usable, 7909.74 MB loaded, full load: True
gguf qtypes: F32 (1087), BF16 (6), Q5_K (146), Q4_K (694)
model weight dtype torch.float16, manual cast: None
model_type FLUX
Requested to load QwenImage
loaded completely; 13754.48 MB usable, 11746.73 MB loaded, full load: True
/workspace/ComfyUI/comfy/samplers.py:966: UserWarning: backend:cudaMallocAsync ignores max_split_size_mb,roundup_power2_divisions, and garbage_collect_threshold. (Triggered internally at /pytorch/c10/cuda/CUDAAllocatorConfig.cpp:391.)
if latent_image is not None and torch.count_nonzero(latent_image) > 0: #Don't shift the empty latent image.
100%|██████████████████████████████████████████████████████████████████████| 9/9 [01:27<00:00, 9.76s/it]
Requested to load WanVAE
0 models unloaded.
loaded partially; 128.00 MB usable, 128.00 MB loaded, 114.00 MB offloaded, lowvram patches: 0
/workspace/ComfyUI/nodes.py:1594: RuntimeWarning: invalid value encountered in cast
img = Image.fromarray(np.clip(i, 0, 255).astype(np.uint8))
Prompt executed in 177.42 seconds
got prompt
Requested to load QwenImage
loaded completely; 13748.48 MB usable, 11746.73 MB loaded, full load: True
100%|██████████████████████████████████████████████████████████████████████| 9/9 [01:28<00:00, 9.88s/it]
Requested to load WanVAE
0 models unloaded.
loaded partially; 128.00 MB usable, 128.00 MB loaded, 114.00 MB offloaded, lowvram patches: 0
Prompt executed in 104.41 seconds
got prompt
Requested to load QwenImage
loaded completely; 13748.48 MB usable, 11746.73 MB loaded, full load: True
85%|█████████████████████████████████████████████████████████▊ | 17/20 [04:44<00:51, 17.06s/it]
Logs
Other
No response