Codebase QA on SWE-Atlas
32.25ScoreLARGERFixed
Evaluation results
| Method | Links | |
|---|---|---|
| LARGERFixedBackbone LLM=GPT-5.22026.05 | 32.25 | |
| Claude Code*Backbone LLM=Claude Opus 4.62026.05 | 31.2 | |
| CodexBackbone LLM=GPT-5.22026.05 | 29.83 | |
| mini-swe-agentBackbone LLM=GPT-5.22026.05 | 25.81 |