agent evals
evals.zndx.orgcoding-agent harness leaderboard
| # | harness | release | pass rate | trials | model | updated |
|---|---|---|---|---|---|---|
| 1 | brisk-kestrel | acp-eval-quick | 100.0% | 3 | Qwen3.6-35B-A3B-FP8 | 117d ago |
| 2 | dapper-falcon | v0.1.2-quick | 100.0% | 3 | Qwen3.6-35B-A3B-FP8 | 117d ago |
| 3 | gilded-ocelot | acp-eval | 100.0% | 30 | Qwen3.8-27B | 36d ago |