Benchmarks · reasoning
FrontierCS 2
Uses the published pass rate against the human reference on every hidden metric, not the separate FCI index. Score: pass rate beating the human reference on every metric. Collected by Jake Boggs' benchmark dashboard.
11 models · 11 results · data as of 2026-10-29 · frontier-cs.org ↗
Leaderboard
Opus 5.5 max
All results
11 rows| # | Model | Effort | FrontierCS 2 |
|---|---|---|---|
| 1 | Opus 5.5 | max | 58.3% |
| 2 | GPT-6 Astra | max | 43.8% |
| 3 | Gemini 4 Argon | max | 39.6% |
| 4 | GPT-6.1 Sol | max | 35.4% |
| 5 | GLM 5.3 | max | 25.0% |
| 6 | Muse Wonder (Max) | max | 20.8% |
| 7 | Kimi K3 | max | 20.8% |
| 8 | DeepSeek V4.1 Flash | max | 14.6% |
| 9 | GLM 5.3 Flash | max | 12.5% |
| 10 | Muse Spark 1.3 | max | 12.5% |
| 11 | Qwen 3.8 Max | max | 12.5% |