如题

I have this error when trying to run [Stable Cascade

I'm using Aki comfyui latest version, I cannot load ur workflow(json)

click load nothing change, no error display, no workflow display, nothing.(loading other workflow json file, works fine)

anyone could fix this?

!!! Exception during processing!!! 'qwen2'
Traceback (most recent call last):
File "F:\install\ComfyUI_windows_portable\ComfyUI\execution.py", line 151, in recursive_execute
output_data, output_ui = get_output_data(obj, input_data_all)
File "F:\install\ComfyUI_windows_portable\ComfyUI\execution.py", line 81, in get_output_data
return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
File "F:\install\ComfyUI_windows_portable\ComfyUI\execution.py", line 74, in map_node_over_list
results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
File "F:\install\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Qwen-2\Qwen2.py", line 27, in load_model
model = AutoModelForCausalLM.from_pretrained(
File "F:\install\ComfyUI_windows_portable\python_embeded\lib\site-packages\transformers\models\auto\auto_factory.py", line 461, in from_pretrained
config, kwargs = AutoConfig.from_pretrained(
File "F:\install\ComfyUI_windows_portable\python_embeded\lib\site-packages\transformers\models\auto\configuration_auto.py", line 998, in from_pretrained
config_class = CONFIG_MAPPING[config_dict["model_type"]]
File "F:\install\ComfyUI_windows_portable\python_embeded\lib\site-packages\transformers\models\auto\configuration_auto.py", line 710, in getitem
raise KeyError(key)
KeyError: 'qwen2'

Error occurred when executing unCLIPCheckpointLoader:

'model.diffusion_model.input_blocks.0.0.weight'

Thank you so much for sharing this!!!

Could you make workflow for the super resolution model?

https://huggingface.co/stabilityai/stable-cascade/blob/main/controlnet/super_resolution.safetensors

Can't find a way to recreate such workflow from Automatic1111:
1 step: generating images with adding some (2-3) additional LORAs. They should be easy switching\adding on\off.
2 step: sending generated image to img2img INPAINT for correction and adding details. Same requirement for LORAs
3 step: sending after step 2 into img2img for custom UPSCALING by X or to custom resolution.
4 step: sending again into INPAINT.

Is it even possible to have all that steps in one workflow and be able to freely switch between it?

dont manage to get
ComfyUI-Flowty-TripoSR-ZHO
running :

ImportError: cannot import name 'remove_background' from 'tsr.utils' (C:\ComfyUI\custom_nodes\ComfyUI-Flowty-TripoSR\tsr\utils.py)

Cannot import C:\ComfyUI\custom_nodes\ComfyUI-Flowty-TripoSR-ZHO module for custom nodes: cannot import name 'remove_background' from 'tsr.utils' (C:\ComfyUI\custom_nodes\ComfyUI-Flowty-TripoSR\tsr\utils.py)

i also can only english , makes it difficult ;)

Where can I find the scinpainting.safetensors model?

生成的3d模型可以导出吗？比如stl格式