SWE-agent takes a GitHub issue and tries to automatically fix it, using GPT-4, or your LM of choice. It can also be employed for offensive cybersecurity or competitive coding challenges.
It slightly lowers performance compared to GPT-4 Turbo, we'll post results to swebench.com in the next few days. Follow one of us on Twitter to get notified.
Describe the issue
What is the effect of using gpt-4o?
Optional: Relevant documentation page
No response