SWE-Gym: An Open Environment for Training Software Engineering Agents and Verifiers
UC Berkeley + UIUC open environment of 2,438 Python repo issues with executable tests for training and verifying SWE-bench-style agents.
- Publisher
- University of California, Berkeley
- Year
- 2024
- Venue
- preprint
- Authors
- 7
- Hosting
- External sourcelicense unknown
Cite
Notes
Only stored in your browser.
Introduces 2 artifacts - 1 eval, 1 tool
TL;DR
Semantic Scholar
This work presents SWE-Gym, the first environment for training real-world software engineering (SWE) agents, and uses SWE-Gym to train language model based SWE agents, achieving up to 19% absolute gains in resolve rate on the popular SWE-Bench Verified and Lite test sets.