Invalid credentials in Authorization header - HfApiModel

[DivyanshJain]

Dear Team,

I am currently enrolled in the Agents course and encountered an error while attempting to run the Smol_Agents notebook. Specifically, I received an error when using model=HfApiModel() with my free Hugging Face API key.

Could you please assist me in resolving this issue?
++++++++++++++++++++++++++++++++++++++++++++++++++
Error:
Error in generating model output:
401 Client Error: Unauthorized for url:
router.huggingface.co/hf-inference/models/Qwen/Qwen2.5-Coder-32B-Instruct/v1/chat/completions (Request ID:
Root=1-67dfa1c6-189363a609d7874972942ec5;5bbc6b0b-6f48-455e-b448-e4d5fc570b68)

Invalid credentials in Authorization header
[Step 1: Duration 2.02 seconds]

HTTPError Traceback (most recent call last)
/usr/local/lib/python3.11/dist-packages/huggingface_hub/utils/_http.py in hf_raise_for_status(response, endpoint_name)
408 try:
→ 409 response.raise_for_status()
410 except HTTPError as e:

11 frames
401 Client Error: Unauthorized for url: https://router.huggingface.co/hf-inference/models/Qwen/Qwen2.5-Coder-32B-Instruct/v1/chat/completions

The above exception was the direct cause of the following exception:

AgentGenerationError Traceback (most recent call last)
/usr/local/lib/python3.11/dist-packages/smolagents/agents.py in step(self, memory_step)
1200 memory_step.model_output = model_output
1201 except Exception as e:
→ 1202 raise AgentGenerationError(f"Error in generating model output:\n{e}", self.logger) from e
1203
1204 self.logger.log_markdown(

Invalid credentials in Authorization header
++++++++++++++++++++++++++++++++++++++++++++++++

[John6666]

The following basic measures may help, but the Inference API is currently unstable and Qwen 2.5 Coder 32B is busy, so there is also the possibility of individual problems…
I’ll bring it if I find something.

Acces token dont work

i can login using cmd like i always get that my token is wrong but i create 3 tokens 3 were wrong i tried everything can someone help?

401 Error: Invalid Credentials in Authorization Header (Flan-T5-Large Inference API)

I have already done the following: Created an API token on Hugging Face Checked that my token has “Make calls to inference providers” permission Tried regenerating a new token and replacing it in my script Confirmed that my Hugging Face account is active and not restricted This is the error message i get: I have already done the following: This is the error message i get: import requests HF_TOKEN = “JOUW_NIEUWE_A…

[DivyanshJain]

Can you confirm if accessing Inference API requires Pro subscription or Free token will work?

[John6666]

I can’t try the Free token at the moment…

[John6666]

The same problem was reported quite a lot in the agents course channel on HF Discord. It would be easier if the following method was still available…

from Discord:

You can use https://pflgm2locj2t89co.us-east-1.aws.endpoints.huggingface.cloud/ as model_id (Qwen 2.5 Coder 32B)