Feature/keywordsai llm

[jordanparker6]

Description

Due to some of the model validation of the LLMs, LLM proxy's that support an OpenAI like API like KeywordsAI required hacky solutions to override the error message when wrapped in OpenAI LLMs.

Currently in discussions with the founders of KeywordsAI to get a review of this so they are happy with the API.

Fixes # (issue)

This provides a KeywordsAI LLM provider that dynamically fetches the model configurations and allows for model params.

New Package?

Did I fill in the tool.llamahub section in the pyproject.toml and provide a detailed README.md for my new integration or package?

• Yes
• No

Version Bump?

Did I bump the version in the pyproject.toml file of the package I am updating? (Except for the llama-index-core package)

• Yes
• No

Type of Change

Please delete options that are not relevant.

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• This change requires a documentation update

How Has This Been Tested?

Your pull-request will likely not be merged unless it is covered by some form of impactful unit testing.

• I added new unit tests to cover this change
• I believe this change is already covered by existing unit tests

Suggested Checklist:

• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• I have added Google Colab support for the newly added notebooks.
• My changes generate no new warnings
• I have added tests that prove my fix is effective or that my feature works
• New and existing unit tests pass locally with my changes
• I ran make format; make lint to appease the lint gods

[jordanparker6]

@logan-markewich just a heads up.

Looking to get a review on this by the founders of keywords to ensure they are happy with the API.

https://www.keywordsai.co/

[logan-markewich]

High-level -- users should use OpenAILike to access openai-like APIs. This would probably avoid some of the checks/errors that you noted in the PR description

That being said, happy to have a specific integration for keywords ai. The implementation likely can be a lot smaller though, most methods are copies from the openai class right?

[logan-markewich]

@jordanparker6 is this good to merge? (let me know if so, it seeeeeems ok?)

[jordanparker6]

@logan-markewich let me quickly ping the keywords guys. There is an opportunity to add a better dev-x here for some of their features (e.g. adding a Customer ID / Thread ID to the LLM logs for observability in their UI). Shouldn't be a big add, just moving the params up to the model params rather than into the metadata param.

[jordanparker6]

@logan-markewich this is good to go