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
11.97k stars 1.2k forks source link

Support Gab.ai API tokens #169

Closed OrcVole closed 3 months ago

OrcVole commented 3 months ago

Describe the feature

OpenAI and Anthropic API tokens are already supported. Please also include support for e.g. Grok and Gab.ai (Arya)

Potential Solutions

Create a generic API token section which can (using examples provided) with mimal edits be used to support AI from other AI companies, or self-hosted AI running at a home-office.

klieret commented 3 months ago

There are PRs for groq #108 #134.

Generally, only very big models show useful performance with SWE-agent.

Feel free to open a PR if you really want to have a certain model supported (though it might make more sense to work on #64), but I will close this as an issue.