AssertionError: You do not have CLIP state dict!
Hello,
I tried to run the stable-diffusion-3.5-large under stable-diffusion-webui-forge but received the error:
"stable-diffusion-webui-forge/backend/loader.py", line 62, in load_huggingface_component
assert isinstance(state_dict, dict) and len(state_dict) > 16, 'You do not have CLIP state dict!'
AssertionError: You do not have CLIP state dict!
You do not have CLIP state dict!
"
I checked the architecture diagram — it's very helpful for understanding the structure of Stable Diffusion 3.5 (SD 3.5), especially the MM-DiT block layout and the triple text encoder input.
No matter that I tried to use text_encoder/model.safetensors + clip_l.safetensors from the Flux repo, the error exists.
Can you tell me how to fix it?
Sorry I can't help, but I have the same problem.
did you ever resolve this issue? I am having the same problem.