AI benchmarks and leaderboards are different things
Learn the difference between a benchmark, an evaluation run and a leaderboard, and use each to answer the right question.
Model reports
SWE Atlas
Reports three splits (Codebase Q&A, Test Writing, Refactoring) as separate figures; the split is part of the instrument.
Evaluation context and evidence
- Source
- Model reports
- Publisher
- Not recorded
- Benchmark release
- 2026-05-08
- Collected
- Not recorded
- Modality
- Not recorded
These are source-reported observations, not independent reruns by this site. Check the task version, evaluation protocol, unit and reporting date before comparing models. Results from different setups are not a universal ranking.
Original sources
Reported results 3
Shown in source observation order, not ranked. Dates retain the precision provided by the source.
| Model | Value | Unit | Protocol | Reported date and basis | Source |
|---|---|---|---|---|---|
| Hy4 preview | 64.0 | percent | Codebase Q&A split | 2026-08-28 · document_publication | Hy4 preview |
| Hy4 preview | 53.3 | percent | Refactoring split | 2026-08-28 · document_publication | Hy4 preview |
| Hy4 preview | 57.8 | percent | Test Writing split | 2026-08-28 · document_publication | Hy4 preview |