SWE-RPG Benchmark
Explore score history ↗A repository-level dataset and protocol evaluating issue resolution through requirement clarification, implementation planning, code generation, and artifact submission, with 163 tasks across 31 Python and Java repositories and validated intermediate ground-truth references.
- Released
- 2026-08-10
- Institutions
- Singapore Management University · Monash University Malaysia · Daegu Gyeongbuk Institute of Science and Technology
- Evaluates
- Code, Code & Software, Code generation, Planning, Software & AI Compute
- Openness
- unknown
- Importer
- Claire Radar
- Review status
- ai-reviewed
SWE-RPG paper and code
No reported scores are on record for this benchmark yet.