When using the azure GPT endpoint, the --model CLI arg is ignored

princeton-nlp / SWE-agent

SWE-agent takes a GitHub issue and tries to automatically fix it, using GPT-4, or your LM of choice. It solves 12.47% of bugs in the SWE-bench evaluation set and takes just 1 minute to run.

https://princeton-nlp.github.io/SWE-agent/

MIT License

11.88k stars 1.19k forks source link

When using the azure GPT endpoint, the --model CLI arg is ignored #571

Open ofirpress opened 3 weeks ago

ofirpress commented 3 weeks ago

hi! When using the azure GPT endpoint, the --model CLI arg is ignored

So we should give a warning (or an error?) if the user tries using it.

(we instead get the model from: OPENAI_API_VERSION: '...') see https://github.com/princeton-nlp/SWE-agent/blob/main/sweagent/agent/models.py#L247

klieret commented 2 weeks ago

(small correction: we get the model from AZURE_OPENAI_DEPLOYMENT)

Yup, this is not ideal. I remember this came up during the review of the PR. Let me revisit if we can't do better at some point (so leaving this open for now)