LTXVPromptEnhancerLoader 401 Client Error - Invalid credentials in Authorization header

[Kozatiju]

Hi ! While using the ltxvideo-i2v json file on comfyui, I have this issue when loading the LTXVPromptEnhancerLoader to download the LLM unsloth/Llama-3.2-3B-Instruct / MiaoshouAI/Florence-2-large-PromptGen-v2.0 :

LTXVPromptEnhancerLoader 401 Client Error. (Request ID: Root=1-67ca14e8-3ed99e584b2f5a392876a0fe;cf22a6f6-6dec-446b-950a-f85868b0d474) Repository Not Found for url: https://huggingface.co/api/models/unsloth/Llama-3.2-3B-Instruct/revision/main. Please make sure you specified the correct repo_id and repo_type. If you are trying to access a private or gated repo, make sure you are authenticated. Invalid credentials in Authorization header

The issue is clearly related to an access not granted because I am not logged. I was wondering where should I enter my huggingface token or log ? Would you have any idea how to fix this issue ?

Thanks for your cool set of nodes for comfyui, they kick ass !!!

[itistripleR]

Try Gemini!

https://www.patreon.com/posts/ltx-gemini-123897759

[able2608]

LTXVPromptEnhancerLoader 401 Client Error. (Request ID: Root=1-67ca14e8-3ed99e584b2f5a392876a0fe;cf22a6f6-6dec-446b-950a-f85868b0d474) Repository Not Found for url: https://huggingface.co/api/models/unsloth/Llama-3.2-3B-Instruct/revision/main. Please make sure you specified the correct repo_id and repo_type. If you are trying to access a private or gated repo, make sure you are authenticated. Invalid credentials in Authorization header

You are not logged in in huggingface-cli. The devs should probably warn about that somewhere, but here we are. For now you'll need to do the following manually (can be done even with ComfyUI launched, will work immediately after completing the steps):

1. Log in to https://huggingface.co/ (If you do not have an account, create one. It's free).
2. Click on your profile picture (on the top right corner) and go to access token page (It's down the list that just popped open).
3. Click the Create new token button on the right. Choose the Read option on the top (the others are fine grained and write), give the token a name, and copy the token to somewhere. You'll not be able to copy the token again. If you do not save it, you'll have to invalidate it and get a new one the next time you need it!
4. Run huggingface-cli login command as follows: "[Your ComfyUI installation path]\python_embeded\Scripts\huggingface-cli" login", then paste in the token you just copied when prompted to do so (you'll not see the content you pasted in and it's totally normal)
5. (Optional) Enter Y when prompt whether to save the access information to keychain.

After that your workflow should be able to download the models required.