Computer Science Problem Solving on FrontierCS
19.82Avg@5FrontierSmith (200)
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| FrontierSmith (200)Model=Qwen3.5-27B2026.05 | 19.82 | 29.38 | |
| FrontierCS (172)Model=Qwen3.5-27B2026.05 | 13.98 | 21.92 | |
| HardTests (200)Model=Qwen3.5-27B2026.05 | 11.2 | 18.37 | |
| FrontierCS (172)Model=Qwen3.5-9B2026.05 | 11.17 | 16.29 | |
| ALE-bench (40)Model=Qwen3.5-9B2026.05 | 10.64 | 15.7 | |
| FrontierSmith (200)Model=Qwen3.5-9B2026.05 | 10.62 | 15.73 | |
| BaseModel=Qwen3.5-27B2026.05 | 7.7 | 13.91 | |
| HardTests (200)Model=Qwen3.5-9B2026.05 | 5.38 | 11.19 | |
| Random Reward (172)Model=Qwen3.5-9B2026.05 | 3.04 | 6.93 | |
| BaseModel=Qwen3.5-9B2026.05 | 1.8 | 5 |