It's not working for me - what do I need to do?

[Rolf-100R]

File "F:\FLUX 2\webui\backend\loader.py", line 61, 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!
Loading Model: {'checkpoint_info': {'filename': 'F:\FLUX 2\webui\models\Stable-diffusion\chroma_v35DetailCalibrated.safetensors', 'hash': '07a82cde'}, 'additional_modules': ['F:\FLUX 2\webui\models\text_encoder\t5xxl_fp8_e4m3fn.safetensors', 'F:\FLUX 2\webui\models\VAE\ae.safetensors', 'F:\FLUX 2\webui\models\text_encoder\diffusion_pytorch_model.safetensors'], 'unet_storage_dtype': None}
StateDict Keys: {'transformer': 643, 'vae': 244, 'text_encoder_2': 220, 'ignore': 0}
Traceback (most recent call last):
File "F:\FLUX 2\webui\modules_forge\main_thread.py", line 30, in work
self.result = self.func(*self.args, **self.kwargs)
File "F:\FLUX 2\webui\modules\txt2img.py", line 131, in txt2img_function
processed = processing.process_images(p)
File "F:\FLUX 2\webui\modules\processing.py", line 836, in process_images
manage_model_and_prompt_cache(p)
File "F:\FLUX 2\webui\modules\processing.py", line 804, in manage_model_and_prompt_cache
p.sd_model, just_reloaded = forge_model_reload()
File "F:\FLUX 2\system\python\lib\site-packages\torch\utils_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "F:\FLUX 2\webui\modules\sd_models.py", line 504, in forge_model_reload
sd_model = forge_loader(state_dict, additional_state_dicts=additional_state_dicts)
File "F:\FLUX 2\system\python\lib\site-packages\torch\utils_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "F:\FLUX 2\webui\backend\loader.py", line 287, in forge_loader
component = load_huggingface_component(estimated_config, component_name, lib_name, cls_name, local_path, component_sd)
File "F:\FLUX 2\webui\backend\loader.py", line 61, 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!

[Rolf-100R]

Installed SD WebUI Forge again, from scratch, and everything seemed to work. But!

1. The picture quality is terrible. Can not achieve anything that looks like photorealism. The picture all the time turns out unnatural, like a cartoon.

2. When generating periodically began to have problems with the computer - first the monitor goes out, and the computer itself can not be restarted even through Reset, only disconnecting from the power.

I have a second SD WebUI Forge on another hard disk, where Flux is running on sampler [Forge] Flux Realistic (2x Slow). And the picture quality is good and there are no problems with the computer.

[maybleMyers]

Use only one vae. ie only use ae.safetensors, I'm not sure it matters but you have two vae selected. Also you need to use like euler simple and do like 36 steps.

[Rolf-100R]

Use only one vae. ie only use ae.safetensors,

Thank you very much! I did it according to your advice.

Can you please tell me if lora works on chromaforge? I really want to get rid of this “cartoonishness” of the image - maybe lora can help.

[maybleMyers]

some loras do. http://discord.gg/SQVcWVbqKx is a good place to look for tips and tricks.

[Rolf-100R]

http://discord.gg/SQVcWVbqKx

Thanks again!