写在最前，感谢香农，感谢所有训练者，数学是穷人的核武器
对AIMETATRON的 红工 |红潮 进行了分层量化 因为我只下了这一个krea2
保留大部分质量的同时立省百分之五十显存
16GB显存畅玩2k 8G畅玩1k
非推理时占用：
2k图片首次推理时占用（LLM未卸载）：
热加载完成后再次推理（LLM卸载后）
对模型的Attention wv（注意力层）MLP down（MLP缩放层）和其他微型层（norm, modulation, embedding ）进行了Q8量化的保留，对【wq, wk, wo（q、k、o）, gate, mlp.gate, mlp.up】等巨型层进行了激进的Q2量化缩放，效果远远大于更大尺寸的Q4量化，且不依赖NVFP4对A卡友好。
原理公式：
注意力机制中（QKV）v对于每次传播系数的影响最大，在llm中v在每次对话中累计其计算复杂度为logN2，但在DIT架构中，由于都是一次性生图/视频，完全没有多轮对话这回事，其计算复杂度完全可以视为2N，所以这里保留了V的的权重系数。
而mlpdown同理他几乎不参与加权计算而作为最终乘数，保留其权重作为残差连接的最后一层，在解码时保留了最终的画质。
从 F16 / Q4_K_S / 混合版 GGUF 中逐 tensor 反量化，对 block 0/7/14/21/27 五层采样：
经架构放大系数 A计算后
误差到输出的传播增益：
wq/wk：A = 0.022（精确值）。误差进 logits 后乘 1/d=1/128=0.088，再过 softmax（雅可比 ≤0.25）：0.088×0.25=0.022，衰减 45 倍
wv / mlp.down：A = 1.0。
wo / gate / up：A ≈ 0.08~0.12（经验标定，与 FLUX GGUF 配方一致）
最终误差推算为 百分之八到十二
也就是说减少一半多的体积只减少至多百分之十二的画质
输出端有效误差最大值（python验算）
最终生图效果对比（相同工作流、提示词、随机种子、生成步数）
图片实测 ：造像
FP8/INT8：
Q2精度混合：
Q4量化：
图片实测 ：人体（多人）
FP8/INT8：
Q4：
Q2混合精度：
复杂场景
int8/q8
Q2混合精度：
Q4：
可以明显注意到，人体错误相对于Q4的减小乃至于完全消失，光影更符合人类逻辑，但物体细节尤其是织物细节/花纹，相对于Q4依旧是缺乏的，当然~可以配合seedvr2.5使用~
请多多反馈意见！可以期待更新BF16-Q2的混合精度版本~因为小水管下载BF16太慢了，工具链还在完善，完成后放出完整量化工具链，明天把纯Q2传了，Q2的意义在于使用更多lora
Description
修复了模型文件一个小维度被压扁的的问题 没有其他改动 llama.cpp误我QAQ
FAQ
Comments (17)
哥，你是把红兽做了个Q2和Q4的量化版本是吗
Actually, this model uses Q2+Q8, and it outperforms Q4.
error load any guff node(
Can you send me the error log, please?：）
@huangzz1234705 later,any guff loader node in comfyui latest version-error with this model
This has been fixed. Please re-download the model file.
@huangzz1234705 no,no work
@artemzlykin369 We have tested this model on five different GPUs – RTX 3090, A100, H100, RX 9070 XT, and RX 580 – and it works fine on all of them. Could you please share your complete error log so we can investigate the issue on your side?
@huangzz1234705
It works even on the latest version of ComfyUI, but only with this package: https://github.com/RealRebelAI/ComfyUI-GGUF_KREA-2 They could have at least written that for us mere mortals—or even just tested it on a clean build; is that really so hard?
@artemzlykin369 I think I know what's going on – you haven't used a Krea‑2 GGUF before, right? The official ComfyUI GGUF nodes don't support Krea‑2. The loader just doesn't recognize its metadata. This isn't a model file issue; it's a framework limitation.
Unet Loader (GGUF) failed?
Can you send me the error log, please?
Use node:ComfyUI-GGUF_KREA-2
Node threw an error during execution.
# ComfyUI Error Report
## Error Details
- Node ID: 363
- Node Type: UnetLoaderGGUF
- Exception Type: IndexError
- Exception Message: IndexError: tuple index out of range
## Stack Trace
```
File "C:\SET UP\ComfyUI_windows_portable\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:\SET UP\ComfyUI_windows_portable\ComfyUI\execution.py", line 344, in get_output_data
return_values = await asyncmap_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:\SET UP\ComfyUI_windows_portable\ComfyUI\execution.py", line 318, in asyncmap_node_over_list
await process_inputs(input_dict, i)
File "C:\SET UP\ComfyUI_windows_portable\ComfyUI\execution.py", line 306, in process_inputs
result = f(**inputs)
File "C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-GGUF_KREA-2\nodes.py", line 176, in load_unet
model = comfy.sd.load_diffusion_model_state_dict(
sd, model_options={"custom_operations": ops}, **kwargs,
)
File "C:\SET UP\ComfyUI_windows_portable\ComfyUI\comfy\sd.py", line 2298, in load_diffusion_model_state_dict
model_config = model_detection.model_config_from_unet(sd, "", metadata=metadata)
File "C:\SET UP\ComfyUI_windows_portable\ComfyUI\comfy\model_detection.py", line 1284, in model_config_from_unet
unet_config = detect_unet_config(state_dict, unet_key_prefix, metadata=metadata)
File "C:\SET UP\ComfyUI_windows_portable\ComfyUI\comfy\model_detection.py", line 981, in detect_unet_config
dit_config["txtlayers"] = state_dict['{}txtfusion.projector.weight'.format(key_prefix)].shape[1]
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^
```
## System Information
- ComfyUI Version: 0.34.3
- Arguments: ComfyUI\main.py --windows-standalone-build --preview-size 256 --reserve-vram 1.0 --highvram
- OS: win32
- Python Version: 3.13.14 (tags/v3.13.14:fd17997, Jun 10 2026, 13:03:48) [MSC v.1944 64 bit (AMD64)]
- Embedded Python: true
- PyTorch Version: 2.13.0+cu130
## Devices
- Name: cuda:0 NVIDIA GeForce RTX 4060 Ti : cudaMallocAsync
- Type: cuda
- VRAM Total: 17175150592
- VRAM Free: 15975055360
- Torch VRAM Total: 0
- Torch VRAM Free: 0
## Logs
```
2026-09-03T21:53:23.915278 - [DONE] Security scan2026-09-03T21:53:23.915292 -
2026-09-03T21:53:24.090081 - ## ComfyUI-Manager: installing dependencies done.2026-09-03T21:53:24.090227 -
2026-09-03T21:53:24.090347 - ** ComfyUI startup time:2026-09-03T21:53:24.090463 - 2026-09-03T21:53:24.090582 - 2026-09-03 21:53:24.0902026-09-03T21:53:24.090700 -
2026-09-03T21:53:24.090822 - ** Platform:2026-09-03T21:53:24.090941 - 2026-09-03T21:53:24.091059 - Windows2026-09-03T21:53:24.091178 -
2026-09-03T21:53:24.091250 - ** Python version:2026-09-03T21:53:24.091344 - 2026-09-03T21:53:24.091406 - 3.13.14 (tags/v3.13.14:fd17997, Jun 10 2026, 13:03:48) [MSC v.1944 64 bit (AMD64)]2026-09-03T21:53:24.091497 -
2026-09-03T21:53:24.091586 - ** Python executable:2026-09-03T21:53:24.091676 - 2026-09-03T21:53:24.091754 - C:\SET UP\ComfyUI_windows_portable\python_embeded\python.exe2026-09-03T21:53:24.091825 -
2026-09-03T21:53:24.091903 - ** ComfyUI Path:2026-09-03T21:53:24.091967 - 2026-09-03T21:53:24.092025 - C:\SET UP\ComfyUI_windows_portable\ComfyUI2026-09-03T21:53:24.092085 -
2026-09-03T21:53:24.092145 - ** ComfyUI Base Folder Path:2026-09-03T21:53:24.092203 - 2026-09-03T21:53:24.092260 - C:\SET UP\ComfyUI_windows_portable\ComfyUI2026-09-03T21:53:24.092326 -
2026-09-03T21:53:24.092413 - ** User directory:2026-09-03T21:53:24.092488 - 2026-09-03T21:53:24.092549 - C:\SET UP\ComfyUI_windows_portable\ComfyUI\user2026-09-03T21:53:24.092609 -
2026-09-03T21:53:24.092756 - ** ComfyUI-Manager config path:2026-09-03T21:53:24.092811 - 2026-09-03T21:53:24.092860 - C:\SET UP\ComfyUI_windows_portable\ComfyUI\user\__manager\config.ini2026-09-03T21:53:24.092910 -
2026-09-03T21:53:24.092963 - ** Log path:2026-09-03T21:53:24.093010 - 2026-09-03T21:53:24.093056 - C:\SET UP\ComfyUI_windows_portable\ComfyUI\user\comfyui.log2026-09-03T21:53:24.093105 -
2026-09-03T21:53:25.310388 - [32m[INFO][0m
Prestartup times for custom nodes:
2026-09-03T21:53:25.310562 - [32m[INFO][0m 0.0 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\rgthree-comfy
2026-09-03T21:53:25.310842 - [32m[INFO][0m 0.0 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Easy-Use
2026-09-03T21:53:25.310965 - [32m[INFO][0m 3.4 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-manager
2026-09-03T21:53:25.311074 - [32m[INFO][0m
2026-09-03T21:53:26.454036 - C:\SET UP\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\cuda\__init__.py:64: 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]
2026-09-03T21:53:30.077773 - [32m[INFO][0m Found comfy_kitchen backend hip: {'available': False, 'disabled': False, 'unavailable_reason': 'PyTorch ROCm/HIP runtime not available', 'capabilities': []}
2026-09-03T21:53:30.078063 - [32m[INFO][0m Found comfy_kitchen backend eager: {'available': True, 'disabled': False, 'unavailable_reason': None, 'capabilities': ['adaln', 'apply_rope', 'apply_rope1', 'apply_rope1_', 'apply_rope_', 'apply_rope_split_half', 'apply_rope_split_half1', 'apply_rope_split_half1_', 'apply_rope_split_half_', 'convrot_w4a4_linear', 'dequantize_convrot_w4a4_weight', 'dequantize_int8_convrot_weight', 'dequantize_int8_convrot_weight_dtype', 'dequantize_int8_embedding', 'dequantize_int8_simple', 'dequantize_int8_simple_dtype', 'dequantize_mxfp8', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'dequantize_w4a8_int8_weight', 'gemv_awq_w4a16', 'int8_linear', 'na3d', 'prepare_int4_weight_for_int8_linear', 'quantize_and_rotate_rowwise', 'quantize_convrot_w4a4_weight', 'quantize_int8_convrot_weight', 'quantize_int8_rowwise', 'quantize_int8_tensorwise', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'quantize_svdquant_w4a4', 'quantize_w4a8_int8_weight', 'rms_adaln', 'rms_rope', 'rms_rope1', 'rms_rope1_', 'rms_rope_', 'rms_rope_split_half', 'rms_rope_split_half1', 'rms_rope_split_half1_', 'rms_rope_split_half_', 'rotate_int8_convrot_weight', 'scaled_mm_mxfp8', 'scaled_mm_nvfp4', 'scaled_mm_svdquant_w4a4', 'stochastic_rounding_fp8', 'w4a8_int8_linear']}
2026-09-03T21:53:30.078950 - [32m[INFO][0m Found comfy_kitchen backend triton: {'available': False, 'disabled': True, 'unavailable_reason': "ImportError: No module named 'triton'", 'capabilities': []}
2026-09-03T21:53:30.079096 - [32m[INFO][0m Found comfy_kitchen backend cuda: {'available': True, 'disabled': False, 'unavailable_reason': None, 'capabilities': ['adaln', 'apply_rope', 'apply_rope1', 'apply_rope1_', 'apply_rope_', 'apply_rope_split_half', 'apply_rope_split_half1', 'apply_rope_split_half1_', 'apply_rope_split_half_', 'convrot_w4a4_linear', 'dequantize_convrot_w4a4_weight', 'dequantize_int8_convrot_weight', 'dequantize_int8_convrot_weight_dtype', 'dequantize_int8_simple', 'dequantize_int8_simple_dtype', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'dequantize_w4a8_int8_weight', 'gemv_awq_w4a16', 'int8_linear', 'na3d', 'prepare_int4_weight_for_int8_linear', 'quantize_and_rotate_rowwise', 'quantize_convrot_w4a4_weight', 'quantize_int8_convrot_weight', 'quantize_int8_rowwise', 'quantize_int8_tensorwise', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'quantize_svdquant_w4a4', 'quantize_w4a8_int8_weight', 'rms_adaln', 'rms_rope', 'rms_rope1', 'rms_rope1_', 'rms_rope_', 'rms_rope_split_half', 'rms_rope_split_half1', 'rms_rope_split_half1_', 'rms_rope_split_half_', 'rotate_int8_convrot_weight', 'scaled_mm_nvfp4', 'scaled_mm_svdquant_w4a4', 'stochastic_rounding_fp8', 'w4a8_int8_linear']}
2026-09-03T21:53:30.089219 - [32m[INFO][0m Checkpoint files will always be loaded safely.
2026-09-03T21:53:30.188307 - [32m[INFO][0m Total VRAM 16380 MB, total RAM 16282 MB
2026-09-03T21:53:30.188522 - [32m[INFO][0m pytorch version: 2.13.0+cu130
2026-09-03T21:53:30.189214 - [32m[INFO][0m Set vram state to: HIGH_VRAM
2026-09-03T21:53:30.189418 - [32m[INFO][0m Device: cuda:0 NVIDIA GeForce RTX 4060 Ti : cudaMallocAsync
2026-09-03T21:53:30.203267 - [32m[INFO][0m Using async weight offloading with 2 streams
2026-09-03T21:53:30.205472 - [32m[INFO][0m Enabled pinned memory 6512.0
2026-09-03T21:53:30.251453 - [32m[INFO][0m Using pytorch attention
2026-09-03T21:53:36.273947 - [32m[INFO][0m Python version: 3.13.14 (tags/v3.13.14:fd17997, Jun 10 2026, 13:03:48) [MSC v.1944 64 bit (AMD64)]
2026-09-03T21:53:36.274221 - [32m[INFO][0m ComfyUI version: 0.34.3
2026-09-03T21:53:36.275412 - [32m[INFO][0m comfy-aimdo version: 0.4.15
2026-09-03T21:53:36.276332 - [32m[INFO][0m comfy-kitchen version: 0.2.31
2026-09-03T21:53:36.281904 - [32m[INFO][0m comfyui-frontend-package version: 1.49.6
2026-09-03T21:53:36.282105 - [32m[INFO][0m comfyui-workflow-templates version: 0.11.54
2026-09-03T21:53:36.282569 - [32m[INFO][0m comfyui-embedded-docs version: 0.5.10
2026-09-03T21:53:36.282690 - [32m[INFO][0m comfy-kitchen version: 0.2.31
2026-09-03T21:53:36.282844 - [32m[INFO][0m comfy-aimdo version: 0.4.15
2026-09-03T21:53:36.284524 - [32m[INFO][0m [Prompt Server] web root: C:\SET UP\ComfyUI_windows_portable\python_embeded\Lib\site-packages\comfyui_frontend_package\static
2026-09-03T21:53:36.285692 - [32m[INFO][0m Asset seeder disabled
2026-09-03T21:53:37.342418 - [32m[INFO][0m No OpenGL_accelerate module loaded: No module named 'OpenGL_accelerate'
2026-09-03T21:53:38.607288 - [1m[33m[WARNING][0m Traceback (most recent call last):
File "C:\SET UP\ComfyUI_windows_portable\ComfyUI\nodes.py", line 2263, in load_custom_node
module_spec.loader.exec_module(module)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^
File "<frozen importlib._bootstrap_external>", line 1019, in exec_module
File "<frozen importlib._bootstrap_external>", line 1156, in get_code
File "<frozen importlib._bootstrap_external>", line 1214, in get_data
FileNotFoundError: [Errno 2] No such file or directory: 'C:\\SET UP\\ComfyUI_windows_portable\\ComfyUI\\custom_nodes\\AAA\\__init__.py'
2026-09-03T21:53:38.607673 - [1m[33m[WARNING][0m Cannot import C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\AAA module for custom nodes: [Errno 2] No such file or directory: 'C:\\SET UP\\ComfyUI_windows_portable\\ComfyUI\\custom_nodes\\AAA\\__init__.py'
2026-09-03T21:53:38.914401 - [Crystools [0;32mINFO[0m] Crystools version: 1.27.4
2026-09-03T21:53:38.950832 - [Crystools [0;32mINFO[0m] Platform release: 10
2026-09-03T21:53:38.951028 - [Crystools [0;32mINFO[0m] JETSON: Not detected.
2026-09-03T21:53:38.952415 - [Crystools [0;32mINFO[0m] CPU: Intel(R) Core(TM) i5-10400F CPU @ 2.90GHz - Arch: AMD64 - OS: Windows 10
2026-09-03T21:53:38.975048 - [Crystools [0;32mINFO[0m] pynvml (NVIDIA) initialized.
2026-09-03T21:53:38.975632 - [Crystools [0;32mINFO[0m] GPU/s:
2026-09-03T21:53:38.991692 - [Crystools [0;32mINFO[0m] 0) NVIDIA GeForce RTX 4060 Ti
2026-09-03T21:53:38.993288 - [Crystools [0;32mINFO[0m] NVIDIA Driver: 610.62
2026-09-03T21:53:41.142094 - W0903 21:53:41.123000 7952 Lib\site-packages\torch\utils\flop_counter.py:29] triton not found; flop counting will not work for triton kernels
2026-09-03T21:53:44.763403 - [ComfyUI-Easy-Use] server: 2026-09-03T21:53:44.763514 - v1.3.7 2026-09-03T21:53:44.763593 - Loaded2026-09-03T21:53:44.763660 -
2026-09-03T21:53:44.763766 - [ComfyUI-Easy-Use] web root: 2026-09-03T21:53:44.764019 - C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Easy-Use\web_version/v2 2026-09-03T21:53:44.764101 - Loaded2026-09-03T21:53:44.764165 -
2026-09-03T21:53:44.776149 - [32m[INFO][0m ComfyUI-GGUF: Allowing full torch compile
2026-09-03T21:53:44.786508 - [32m[INFO][0m ComfyUI-GGUF: Allowing full torch compile
2026-09-03T21:53:44.814877 - [CCTech Suite]: 🤖🤖🤖 2026-09-03T21:53:44.815073 - Quantized and caffeinated!2026-09-03T21:53:44.815326 - 🤖🤖🤖2026-09-03T21:53:44.815422 -
2026-09-03T21:53:44.815509 - [CCTech Suite]:2026-09-03T21:53:44.815774 - Activated 2026-09-03T21:53:44.815890 - 732026-09-03T21:53:44.815960 - GGUF loader nodes (🤖 CCTech/GGUF).2026-09-03T21:53:44.816024 -
2026-09-03T21:53:44.835148 - [32m[INFO][0m ComfyUI-GGUF: Allowing full torch compile
2026-09-03T21:53:44.845715 - [32m[INFO][0m ### Loading: ComfyUI-Impact-Pack (V8.28.3)
2026-09-03T21:53:45.072797 - [32m[INFO][0m [Impact Pack] Wildcard total size (0.00 MB) is within cache limit (50.00 MB). Using full cache mode.
2026-09-03T21:53:45.074315 - [32m[INFO][0m [Impact Pack] Wildcards loading done.
2026-09-03T21:53:45.081474 - [32m[INFO][0m ### Loading: ComfyUI-Impact-Subpack (V1.3.5)
2026-09-03T21:53:45.084901 - [32m[INFO][0m [Impact Pack/Subpack] Using folder_paths to determine whitelist path: C:\SET UP\ComfyUI_windows_portable\ComfyUI\user\default\ComfyUI-Impact-Subpack\model-whitelist.txt
2026-09-03T21:53:45.085528 - [32m[INFO][0m [Impact Pack/Subpack] Ensured whitelist directory exists: C:\SET UP\ComfyUI_windows_portable\ComfyUI\user\default\ComfyUI-Impact-Subpack
2026-09-03T21:53:45.085871 - [32m[INFO][0m [Impact Pack/Subpack] Loaded 0 model(s) from whitelist: C:\SET UP\ComfyUI_windows_portable\ComfyUI\user\default\ComfyUI-Impact-Subpack\model-whitelist.txt
2026-09-03T21:53:45.247007 - [32m[INFO][0m [Impact Subpack] ultralytics_bbox: C:\SET UP\ComfyUI_windows_portable\ComfyUI\models\ultralytics\bbox
2026-09-03T21:53:45.247324 - [32m[INFO][0m [Impact Subpack] ultralytics_segm: C:\SET UP\ComfyUI_windows_portable\ComfyUI\models\ultralytics\segm
2026-09-03T21:53:45.265807 -
================================================================================2026-09-03T21:53:45.265965 -
2026-09-03T21:53:45.266631 - WARNING: llama-cpp-python library not found, GGUF functionality is not available2026-09-03T21:53:45.266930 -
2026-09-03T21:53:45.267035 - To use GGUF features, install additional dependencies:2026-09-03T21:53:45.267174 -
2026-09-03T21:53:45.267265 - pip install -r requirements_gguf.txt2026-09-03T21:53:45.267426 -
2026-09-03T21:53:45.267677 - Basic JoyCaption functionality is still available2026-09-03T21:53:45.267767 -
2026-09-03T21:53:45.267854 - ================================================================================
2026-09-03T21:53:45.268008 -
2026-09-03T21:53:45.284897 - [JoyCaption] ℹ️ No custom models found, skipping user-defined HF models.2026-09-03T21:53:45.285048 -
2026-09-03T21:53:45.303002 - [32m[INFO][0m ### Loading: ComfyUI-Manager (V3.41)
2026-09-03T21:53:45.303775 - [32m[INFO][0m [ComfyUI-Manager] network_mode: public
2026-09-03T21:53:45.303993 - [32m[INFO][0m [ComfyUI-Manager] ComfyUI per-queue preview override detected (PR #11261). Manager's preview method feature is disabled. Use ComfyUI's --preview-method CLI option or 'Settings > Execution > Live preview method'.
2026-09-03T21:53:45.476744 - [32m[INFO][0m ### ComfyUI Revision: 264 [87465b8f] *DETACHED | Released on '2026-09-02'
2026-09-03T21:53:45.556124 - [32m[INFO][0m Model Mover: API routes registered successfully
2026-09-03T21:53:45.600301 - [SD Prompt Reader] 2026-09-03T21:53:45.600416 - Node version: 1.3.42026-09-03T21:53:45.600496 -
2026-09-03T21:53:45.600594 - [SD Prompt Reader] 2026-09-03T21:53:45.600728 - Core version: 1.3.52026-09-03T21:53:45.600794 -
2026-09-03T21:53:45.760214 - [32m[INFO][0m [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/alter-list.json
2026-09-03T21:53:45.878830 - [32m[INFO][0m [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/model-list.json
2026-09-03T21:53:46.239860 - [32m[INFO][0m [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/github-stats.json
2026-09-03T21:53:46.400814 - [32m[INFO][0m [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/extension-node-map.json
2026-09-03T21:53:46.577923 - [32m[INFO][0m [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/custom-node-list.json
2026-09-03T21:53:46.590785 - ------------------------------------------2026-09-03T21:53:46.590892 -
2026-09-03T21:53:46.591070 - Comfyroll Studio v1.76 : 2026-09-03T21:53:46.591254 - 175 Nodes Loaded2026-09-03T21:53:46.591334 -
2026-09-03T21:53:46.591431 - ------------------------------------------2026-09-03T21:53:46.591573 -
2026-09-03T21:53:46.591649 - ** For changes, please see patch notes at https://github.com/Suzie1/ComfyUI_Comfyroll_CustomNodes/blob/main/Patch_Notes.md2026-09-03T21:53:46.591799 -
2026-09-03T21:53:46.591864 - ** For help, please see the wiki at https://github.com/Suzie1/ComfyUI_Comfyroll_CustomNodes/wiki2026-09-03T21:53:46.592045 -
2026-09-03T21:53:46.592121 - ------------------------------------------2026-09-03T21:53:46.592323 -
2026-09-03T21:53:46.612030 - [C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui_controlnet_aux] | INFO -> Using ckpts path: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui_controlnet_aux\ckpts2026-09-03T21:53:46.617146 -
2026-09-03T21:53:46.622284 - [C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui_controlnet_aux] | INFO -> Using symlinks: False2026-09-03T21:53:46.627455 -
2026-09-03T21:53:46.632654 - [C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui_controlnet_aux] | INFO -> Using ort providers: ['CUDAExecutionProvider', 'DirectMLExecutionProvider', 'OpenVINOExecutionProvider', 'ROCMExecutionProvider', 'CPUExecutionProvider', 'CoreMLExecutionProvider']2026-09-03T21:53:46.637863 -
2026-09-03T21:53:46.668053 - C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui_controlnet_aux\node_wrappers\dwpose.py:26: UserWarning: DWPose: Onnxruntime not found or doesn't come with acceleration providers, switch to OpenCV with CPU device. DWPose might run very slowly
warnings.warn("DWPose: Onnxruntime not found or doesn't come with acceleration providers, switch to OpenCV with CPU device. DWPose might run very slowly")
2026-09-03T21:53:47.292229 -
2026-09-03T21:53:47.292341 - Efficiency Nodes:2026-09-03T21:53:47.292422 - Attempting to add Control Net options to the 'HiRes-Fix Script' Node (comfyui_controlnet_aux add-on)...2026-09-03T21:53:47.292489 - Success!2026-09-03T21:53:47.292552 -
2026-09-03T21:53:47.351224 -
2026-09-03T21:53:47.351398 - [rgthree-comfy] Loaded 48 magnificent nodes. 🎉2026-09-03T21:53:47.351504 -
2026-09-03T21:53:47.351697 -
2026-09-03T21:53:47.352014 - [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.
2026-09-03T21:53:47.352497 -
2026-09-03T21:53:47.359125 - [32m[INFO][0m
Import times for custom nodes:
2026-09-03T21:53:47.359355 - [32m[INFO][0m 0.0 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\websocket_image_save.py
2026-09-03T21:53:47.359515 - [32m[INFO][0m 0.0 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui_lora_tag_loader
2026-09-03T21:53:47.359656 - [32m[INFO][0m 0.0 seconds (IMPORT FAILED): C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\AAA
2026-09-03T21:53:47.359834 - [32m[INFO][0m 0.0 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Anima-Enhancer
2026-09-03T21:53:47.359975 - [32m[INFO][0m 0.0 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\lora-info
2026-09-03T21:53:47.360109 - [32m[INFO][0m 0.0 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\joycaption_comfyui
2026-09-03T21:53:47.360232 - [32m[INFO][0m 0.0 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\sd-dynamic-thresholding
2026-09-03T21:53:47.360384 - [32m[INFO][0m 0.0 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Krea2T-Enhancer
2026-09-03T21:53:47.360513 - [32m[INFO][0m 0.0 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\Anima-Mod-Guidance-ComfyUI-Node
2026-09-03T21:53:47.360635 - [32m[INFO][0m 0.0 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-GGUF
2026-09-03T21:53:47.360784 - [32m[INFO][0m 0.0 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Custom-Scripts
2026-09-03T21:53:47.360910 - [32m[INFO][0m 0.0 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-adaptiveprompts
2026-09-03T21:53:47.361032 - [32m[INFO][0m 0.0 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-prompt-control
2026-09-03T21:53:47.361182 - [32m[INFO][0m 0.0 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\gguf
2026-09-03T21:53:47.361305 - [32m[INFO][0m 0.0 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-GGUF_KREA-2
2026-09-03T21:53:47.361424 - [32m[INFO][0m 0.0 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-prompt-reader-node
2026-09-03T21:53:47.361572 - [32m[INFO][0m 0.0 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\rgthree-comfy
2026-09-03T21:53:47.361694 - [32m[INFO][0m 0.0 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-JoyCaption
2026-09-03T21:53:47.361865 - [32m[INFO][0m 0.0 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-GGUF-Loader
2026-09-03T21:53:47.362001 - [32m[INFO][0m 0.0 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Model-Mover
2026-09-03T21:53:47.362125 - [32m[INFO][0m 0.1 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Translator
2026-09-03T21:53:47.362270 - [32m[INFO][0m 0.1 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui_controlnet_aux
2026-09-03T21:53:47.362388 - [32m[INFO][0m 0.2 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-impact-subpack
2026-09-03T21:53:47.362538 - [32m[INFO][0m 0.2 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-manager
2026-09-03T21:53:47.362663 - [32m[INFO][0m 0.2 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Impact-Pack
2026-09-03T21:53:47.362779 - [32m[INFO][0m 0.4 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Crystools
2026-09-03T21:53:47.362947 - [32m[INFO][0m 0.6 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\efficiency-nodes-comfyui
2026-09-03T21:53:47.363250 - [32m[INFO][0m 0.9 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_Comfyroll_CustomNodes
2026-09-03T21:53:47.363517 - [32m[INFO][0m 2.6 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Easy-Use
2026-09-03T21:53:47.363648 - [32m[INFO][0m 3.2 seconds: C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-dynamicprompts
2026-09-03T21:53:47.363895 - [32m[INFO][0m
2026-09-03T21:53:47.367319 - [32m[INFO][0m Context impl SQLiteImpl.
2026-09-03T21:53:47.367480 - [32m[INFO][0m Will assume non-transactional DDL.
2026-09-03T21:53:47.606982 - [32m[INFO][0m Using RAM pressure cache.
2026-09-03T21:53:47.607935 - [32m[INFO][0m Starting server
2026-09-03T21:53:47.608590 - [32m[INFO][0m To see the GUI go to: http://127.0.0.1:8188
2026-09-03T21:53:50.846207 - [Adaptive Prompts] Config Updated: {'default_rng_mode': 'Adaptive'}2026-09-03T21:53:50.846357 -
2026-09-03T21:53:50.847425 - [Adaptive Prompts] Config Updated: {'resolution_strategy': 'Scoped'}2026-09-03T21:53:50.847553 -
2026-09-03T21:53:50.849007 - [Adaptive Prompts] Config Updated: {'search_depth_limit': 80}2026-09-03T21:53:50.849127 -
2026-09-03T21:53:50.849907 - [Adaptive Prompts] Config Updated: {'hide_comments': True}2026-09-03T21:53:50.850013 -
2026-09-03T21:53:50.851055 - [Adaptive Prompts] Config Updated: {'missing_wildcard_behavior': 'Inject Warning'}2026-09-03T21:53:50.851160 -
2026-09-03T21:53:50.851661 - [1m[33m[WARNING][0m [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.
2026-09-03T21:53:50.852053 - [1m[33m[WARNING][0m [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.
2026-09-03T21:53:50.868256 - [1m[33m[WARNING][0m [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.
2026-09-03T21:53:50.868857 - [1m[33m[WARNING][0m [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.
2026-09-03T21:53:53.999834 - [Adaptive Prompts] Config Updated: {'default_rng_mode': 'Adaptive'}2026-09-03T21:53:53.999966 -
2026-09-03T21:53:54.001014 - [Adaptive Prompts] Config Updated: {'resolution_strategy': 'Scoped'}2026-09-03T21:53:54.001119 -
2026-09-03T21:53:54.001922 - [Adaptive Prompts] Config Updated: {'search_depth_limit': 80}2026-09-03T21:53:54.002024 -
2026-09-03T21:53:54.002875 - [Adaptive Prompts] Config Updated: {'hide_comments': True}2026-09-03T21:53:54.002968 -
2026-09-03T21:53:54.003678 - [Adaptive Prompts] Config Updated: {'missing_wildcard_behavior': 'Inject Warning'}2026-09-03T21:53:54.003773 -
2026-09-03T21:53:56.323823 - [Adaptive Prompts] Config Updated: {'default_rng_mode': 'Adaptive'}2026-09-03T21:53:56.323954 -
2026-09-03T21:53:56.324894 - [Adaptive Prompts] Config Updated: {'search_depth_limit': 80}2026-09-03T21:53:56.325007 -
2026-09-03T21:53:56.326034 - [Adaptive Prompts] Config Updated: {'hide_comments': True}2026-09-03T21:53:56.326140 -
2026-09-03T21:53:56.327345 - [Adaptive Prompts] Config Updated: {'resolution_strategy': 'Scoped'}2026-09-03T21:53:56.327498 -
2026-09-03T21:53:56.329213 - [Adaptive Prompts] Config Updated: {'missing_wildcard_behavior': 'Inject Warning'}2026-09-03T21:53:56.329351 -
2026-09-03T21:53:56.335393 - [1m[33m[WARNING][0m [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.
2026-09-03T21:53:57.279656 - [Adaptive Prompts] Config Updated: {'default_rng_mode': 'Adaptive'}2026-09-03T21:53:57.279794 -
2026-09-03T21:53:57.280680 - [Adaptive Prompts] Config Updated: {'search_depth_limit': 80}2026-09-03T21:53:57.280768 -
2026-09-03T21:53:57.281728 - [Adaptive Prompts] Config Updated: {'hide_comments': True}2026-09-03T21:53:57.281847 -
2026-09-03T21:53:57.283420 - [Adaptive Prompts] Config Updated: {'resolution_strategy': 'Scoped'}2026-09-03T21:53:57.283530 -
2026-09-03T21:53:57.284360 - [Adaptive Prompts] Config Updated: {'missing_wildcard_behavior': 'Inject Warning'}2026-09-03T21:53:57.284468 -
2026-09-03T21:54:39.111101 - [Adaptive Prompts] Config Updated: {'default_rng_mode': 'Adaptive'}2026-09-03T21:54:39.111241 -
2026-09-03T21:54:39.112798 - [Adaptive Prompts] Config Updated: {'resolution_strategy': 'Scoped'}2026-09-03T21:54:39.112940 -
2026-09-03T21:54:39.114213 - [Adaptive Prompts] Config Updated: {'search_depth_limit': 80}2026-09-03T21:54:39.114336 -
2026-09-03T21:54:39.115212 - [Adaptive Prompts] Config Updated: {'hide_comments': True}2026-09-03T21:54:39.115326 -
2026-09-03T21:54:39.116253 - [Adaptive Prompts] Config Updated: {'missing_wildcard_behavior': 'Inject Warning'}2026-09-03T21:54:39.116363 -
2026-09-03T21:54:42.052950 - [Adaptive Prompts] Config Updated: {'default_rng_mode': 'Adaptive'}2026-09-03T21:54:42.053080 -
2026-09-03T21:54:42.054236 - [Adaptive Prompts] Config Updated: {'search_depth_limit': 80}2026-09-03T21:54:42.054371 -
2026-09-03T21:54:42.055070 - [Adaptive Prompts] Config Updated: {'hide_comments': True}2026-09-03T21:54:42.055150 -
2026-09-03T21:54:42.055821 - [Adaptive Prompts] Config Updated: {'resolution_strategy': 'Scoped'}2026-09-03T21:54:42.055900 -
2026-09-03T21:54:42.056581 - [Adaptive Prompts] Config Updated: {'missing_wildcard_behavior': 'Inject Warning'}2026-09-03T21:54:42.056664 -
2026-09-03T21:55:22.790005 - [32m[INFO][0m got prompt
2026-09-03T21:55:22.792937 - [1m[31m[ERROR][0m Failed to validate prompt for output 1:
2026-09-03T21:55:22.796593 - [1m[31m[ERROR][0m * LoraTagLoader 126:
2026-09-03T21:55:22.797691 - [1m[31m[ERROR][0m - Required input is missing: clip
2026-09-03T21:55:22.798169 - [1m[31m[ERROR][0m Output will be ignored
2026-09-03T21:55:22.805363 - [1m[31m[ERROR][0m Failed to validate prompt for output 98:
2026-09-03T21:55:22.805524 - [1m[31m[ERROR][0m Output will be ignored
2026-09-03T21:55:22.805864 - [1m[31m[ERROR][0m Failed to validate prompt for output 119:
2026-09-03T21:55:22.806152 - [1m[31m[ERROR][0m Output will be ignored
2026-09-03T21:55:22.999855 - [32m[INFO][0m [32mPrompt executed in 0.19 seconds[0m
2026-09-03T21:56:36.803235 - [32m[INFO][0m got prompt
2026-09-03T21:56:37.030623 - New prompt: She is leaning over a teachers desk with her chest resting on it, legs straight feet flat to the floor. Her palms flat on the desk stretching forward. The man has his left hand pressed against her left buttock, and his right hand pulled back as if about to spank the girls ass. (Her light light brown underwear is pulled down around her ankles:1.1). she has ivory-white skinned, curvy body; She has black long wavy hair and big hair, she has ivory-white skinned, curvy body; She has dark brown short hair and big hair, completely in frame, dressed as a schoolgirl, long white socks, very slim, skinny, cute, small and short, wearing lots of makeup, eyeshadow, mascara and shiny lip gloss. Black high heel shoes. A little bit of rouge on her cheeks. She is looking at the viewer over her right shoulder. She has distressed and lewd expression with Her eyes are half-closed, her mouth is wide opened for shouting, furrowed her eyebrows. We are looking at her from behind and to the side. A man who is mostly out of frame dressed as a teacher is groping one of her ass cheeks, his whole right palm pressed against her, her ass cheeks are ever so slightly pink with a vague hand imprint, her ass is bare and we can see her pussy, no pubic hair. She wears school uniform with white sailor shirt, Her short skirt lifted over her hip, exposed her round, sexy buttocks and her beautiful legs; she has shaved pussy, pussy stubble. The man has his large, long erect penis out. The scene is an old boarding school classroom, dark wooden panelling, natural daylight.,
The photo was taken in an amateur. Highly detailed skin texture, bright, clearly, high contrast, high saturation, vibrance, deep shadows, sharp details, Asian, East Asian, Southeast Asian, lipgloss, glossy skin, tall female.2026-09-03T21:56:37.030923 -
2026-09-03T21:56:37.090621 - [32m[INFO][0m Using pytorch attention in VAE
2026-09-03T21:56:37.092378 - [32m[INFO][0m Using pytorch attention in VAE
2026-09-03T21:56:38.014169 - [32m[INFO][0m VAE load device: cuda:0, offload device: cpu, dtype: torch.bfloat16
2026-09-03T21:56:38.042407 - New prompt: 1024x15362026-09-03T21:56:38.042573 -
2026-09-03T21:56:38.128994 - [32m[INFO][0m Found quantization metadata version 1
2026-09-03T21:56:38.133189 - [32m[INFO][0m Using MixedPrecisionOps for text encoder
2026-09-03T21:56:40.658368 - [32m[INFO][0m Requested to load Krea2TEModel_
2026-09-03T21:56:40.753779 - [32m[INFO][0m loaded completely; 4999.47 MB loaded, full load: True
2026-09-03T21:56:40.779897 - [32m[INFO][0m CLIP/text encoder model load device: cuda:0, offload device: cpu, current: cuda:0, dtype: torch.float16
2026-09-03T21:56:41.138412 - [32m[INFO][0m gguf qtypes: F32 (166), F16 (264), Q2_K (192), Q8_0 (64)
2026-09-03T21:56:41.149229 - [1m[31m[ERROR][0m !!! Exception during processing !!! tuple index out of range
2026-09-03T21:56:41.156204 - [1m[31m[ERROR][0m Traceback (most recent call last):
File "C:\SET UP\ComfyUI_windows_portable\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:\SET UP\ComfyUI_windows_portable\ComfyUI\execution.py", line 344, in get_output_data
return_values = await asyncmap_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:\SET UP\ComfyUI_windows_portable\ComfyUI\execution.py", line 318, in asyncmap_node_over_list
await process_inputs(input_dict, i)
File "C:\SET UP\ComfyUI_windows_portable\ComfyUI\execution.py", line 306, in process_inputs
result = f(**inputs)
File "C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-GGUF_KREA-2\nodes.py", line 176, in load_unet
model = comfy.sd.load_diffusion_model_state_dict(
sd, model_options={"custom_operations": ops}, **kwargs,
)
File "C:\SET UP\ComfyUI_windows_portable\ComfyUI\comfy\sd.py", line 2298, in load_diffusion_model_state_dict
model_config = model_detection.model_config_from_unet(sd, "", metadata=metadata)
File "C:\SET UP\ComfyUI_windows_portable\ComfyUI\comfy\model_detection.py", line 1284, in model_config_from_unet
unet_config = detect_unet_config(state_dict, unet_key_prefix, metadata=metadata)
File "C:\SET UP\ComfyUI_windows_portable\ComfyUI\comfy\model_detection.py", line 981, in detect_unet_config
dit_config["txtlayers"] = state_dict['{}txtfusion.projector.weight'.format(key_prefix)].shape[1]
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^
IndexError: tuple index out of range
2026-09-03T21:56:41.160072 - [32m[INFO][0m [32mPrompt executed in 4.34 seconds[0m
```
## 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)
Use node:ComfyUI-GGUF
[ERROR] !!! Exception during processing !!! tuple index out of range
[ERROR] Traceback (most recent call last):
File "C:\SET UP\ComfyUI_windows_portable\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:\SET UP\ComfyUI_windows_portable\ComfyUI\execution.py", line 344, in get_output_data
return_values = await asyncmap_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:\SET UP\ComfyUI_windows_portable\ComfyUI\execution.py", line 318, in asyncmap_node_over_list
await process_inputs(input_dict, i)
File "C:\SET UP\ComfyUI_windows_portable\ComfyUI\execution.py", line 306, in process_inputs
result = f(**inputs)
File "C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\gguf\pig.py", line 581, in load_model
model = comfy.sd.load_diffusion_model_state_dict(sd, model_options=
{"custom_operations": ops}, **kwargs,)
File "C:\SET UP\ComfyUI_windows_portable\ComfyUI\comfy\sd.py", line 2298, in load_diffusion_model_state_dict
model_config = model_detection.model_config_from_unet(sd, "", metadata=metadata)
File "C:\SET UP\ComfyUI_windows_portable\ComfyUI\comfy\model_detection.py", line 1284, in model_config_from_unet
unet_config = detect_unet_config(state_dict, unet_key_prefix, metadata=metadata)
File "C:\SET UP\ComfyUI_windows_portable\ComfyUI\comfy\model_detection.py", line 981, in detect_unet_config
dit_config["txtlayers"] = state_dict['{}txtfusion.projector.weight'.format(key_prefix)].shape[1]
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^
IndexError: tuple index out of range
Use node:gguf
[ERROR] !!! Exception during processing !!! tuple index out of range
[ERROR] Traceback (most recent call last):
File "C:\SET UP\ComfyUI_windows_portable\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:\SET UP\ComfyUI_windows_portable\ComfyUI\execution.py", line 344, in get_output_data
return_values = await asyncmap_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:\SET UP\ComfyUI_windows_portable\ComfyUI\execution.py", line 318, in asyncmap_node_over_list
await process_inputs(input_dict, i)
File "C:\SET UP\ComfyUI_windows_portable\ComfyUI\execution.py", line 306, in process_inputs
result = f(**inputs)
File "C:\SET UP\ComfyUI_windows_portable\ComfyUI\custom_nodes\gguf\pig.py", line 581, in load_model
model = comfy.sd.load_diffusion_model_state_dict(sd, model_options=
{"custom_operations": ops}, **kwargs,)
File "C:\SET UP\ComfyUI_windows_portable\ComfyUI\comfy\sd.py", line 2298, in load_diffusion_model_state_dict
model_config = model_detection.model_config_from_unet(sd, "", metadata=metadata)
File "C:\SET UP\ComfyUI_windows_portable\ComfyUI\comfy\model_detection.py", line 1284, in model_config_from_unet
unet_config = detect_unet_config(state_dict, unet_key_prefix, metadata=metadata)
File "C:\SET UP\ComfyUI_windows_portable\ComfyUI\comfy\model_detection.py", line 981, in detect_unet_config
dit_config["txtlayers"] = state_dict['{}txtfusion.projector.weight'.format(key_prefix)].shape[1]
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^
IndexError: tuple index out of range
I used three nodes to load UNet models from three different authors, but all of them resulted in errors.
We've identified and fixed the issue. The root cause was that llama.cpp compressed the dimensions in the file header, which passed in the test environment but triggered an error in production due to Python script validation. Please re-download the model file – it should now work properly.