search

All-Hands-AI / OpenHands

🙌 OpenHands: Code Less, Make More
https://all-hands.dev
MIT License
37.37k stars 4.23k forks source link

[Resolver] API Retry on guess success #5187

Open malhotra5 opened 4 days ago

malhotra5 commented 4 days ago

End-user friendly description of the problem this fixes or functionality that this introduces

Give a summary of what the PR does, explaining any non-trivial design decisions

This PR partially addresses the problem of rate limit retries

  1. We've replaced all litellm.completion instances with the LLM class from Openhands proper. This class supports retry with the following configurable parameters:

    • LLM_NUM_RETRIES (Default of 8)
    • LLM_RETRY_MIN_WAIT (Default of 15 seconds)
    • LLM_RETRY_MAX_WAIT (Default of 120 seconds)

    • LLM_RETRY_MULTIPLIER (Default of 2)

      This refactor helps centralize retry mechanism logic and applies it on guess_success in openhands-resolver via the LLM class

  2. Updated test cases appropriately. Also added test cases for retry on rate limit exception for guess_success

Note: while this PR implements retry mechanism it is not a perfect solution for Anthropic's token per minute rate limit exceptions. However, if this PR is merged more sophisticated retry mechanism can be implemented openhands/llm/llm.py and openhands/llm/retry_mixin.py, which will apply to both Openhands proper and Openhands resolver.

Link of any specific issues this addresses

5030

5087