Support gpt 3.5 turbo 16k model

coderabbitai / ai-pr-reviewer

AI-based Pull Request Summarizer and Reviewer with Chat Capabilities.

https://coderabbit.ai

MIT License

1.41k stars 258 forks source link

Support gpt 3.5 turbo 16k model #424

Closed HyunggyuJang closed 11 months ago

HyunggyuJang commented 11 months ago

TokenLimits is only the place needed to be modified. Have set the token limits accordingly. Closes #406

Summary by CodeRabbit

New Feature:

Added support for the "gpt-3.5-turbo-16k" model in the TokenLimits class.
Set the maxTokens limit to 16300 and the responseTokens limit to 3000 for the new model.

🎉 With tokens aplenty, we set the stage, For the "gpt-3.5-turbo-16k" to engage. More power, more wisdom, in every page, A new chapter begins, let's turn the page! 🚀

github-actions[bot] commented 11 months ago

CodeRabbit

Walkthrough

The changes introduced in this pull request primarily focus on extending the functionality of the TokenLimits class. This is achieved by adding support for a new model, "gpt-3.5-turbo-16k". The token limits for this model have been set to ensure optimal performance and compliance with the model's specifications.

Changes

File	Summary
`src/limits.ts`	The `TokenLimits` class has been updated to include support for the "gpt-3.5-turbo-16k" model. The maximum tokens limit has been set to 16300 and the response tokens limit to 3000 for this model.