Small issues when running "Migrating from OpenAI to Open LLMs Using TGI’s Messages API" in Colab

[sergiopaniego]

I tried running Migrating from OpenAI to Open LLMs Using TGI’s Messages API in Colab and found the following issues:

• There is a compatibility issue when doing the upgrade of package versions in the first cell.

  ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
  torchaudio 2.3.1+cu121 requires torch==2.3.1, but you have torch 2.4.0 which is incompatible.
  torchvision 0.18.1+cu121 requires torch==2.3.1, but you have torch 2.4.0 which is incompatible.

• There is an issue with create_inference_endpoint since it looks like it is not taking the API_TOKEN correctly.

  LocalTokenNotFoundError: Token is required (`token=True`), but no token found. You need to provide a token or be logged in to Hugging Face with `huggingface-cli login` or `huggingface_hub.login`. See https://huggingface.co/settings/tokens.

• Llama index has suffered some changes in the API that generate errors when running the code.

  
  ImportError: cannot import name 'OpenAILike' from 'llama_index.llms' (unknown location)

or

ImportError: cannot import name 'ServiceContext' from 'llama_index' (unknown location)