FrontierCode Main: leaderboard
Best published five-trial mean rubric score for each model across tested reasoning efforts on 100 Main tasks; selected effort and evaluation harness are retained.
Metric: Rubric Score (%). Source: cognition.ai. Saturation forecast: Around 2028. 12 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.5 (xHigh) | 25.48 | #31 (GPT-5.5) |
| 2 | GPT-5.4 Mini (xHigh) | 17.82 | #212 (GPT-5.4 Mini) |
| 3 | Kimi K2.6 | 16.04 | #104 |
| 4 | Kimi K2.5 | 6.86 | #144 |
| 5 | MiniMax-M2.7 | 5.98 | #258 |
| 6 | MiniMax-M2.5 | 5.34 | #303 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=frontiercode-main · How It Works · Data refreshed daily, snapshot 2026-10-20.