Persistent 'Not Found' (404) Error on Inference API with New Account & Valid Token

Beginners
1

Hello,

I am a new user trying to use the Inference API. I am consistently receiving a 404 Not Found error for all models, including public ones like gpt2.

I have already tried the following troubleshooting steps:

  • Confirmed I have been granted access to gated models like Llama 3.1 and Gemma.
  • Generated a new, clean, read-only User Access Token.
  • Verified the token is being passed correctly in the Authorization: Bearer <token> header.
  • Confirmed the error still happens with a direct cURL command, even for the public gpt2 model.

Here is the result of my cURL test:

curl https://api-inference.huggingface.co/models/gpt2 -X POST -d '{"inputs": "Hello there"}' -H "Content-Type: application/json" -H "Authorization: Bearer <Your_Token_Here>" Not Found

Could you please check if there is an issue or restriction on my account? My Hugging Face username is abacus808.

Thank you!

1 Like
2

The usage of the Inference API and the range of models that can be used are currently being updated. The endpoint URL and other details are changing. For more information, please refer to the following post or contact Hugging Face. website@huggingface.co feedback@huggingface.co
openai-community/gpt2 · Hugging Face
Persistent 404 Not Found Errors with Public Inference API - #10 by meganariley
Exceeded your monthly included credits for Inference Providers