FrontierCode: leaderboard
Cognition benchmark for production-quality coding agents measuring whether maintainers would merge model PRs. It uses 150 maintainer-authored open source tasks with nested Extended, Main, and Diamond subsets scored by blockers and quality rubrics.
Metric: Rubric Score (%). Source: cognition.ai. 12 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.5 (Medium) | 6.31 | #31 (GPT-5.5) |
| 2 | GPT-5.4 Mini (xHigh) | 4.58 | #212 (GPT-5.4 Mini) |
| 3 | Kimi K2.6 | 3.77 | #104 |
| 4 | MiniMax-M2.7 | 2.38 | #258 |
| 5 | MiniMax-M2.5 | 1.08 | #303 |
| 6 | Kimi K2.5 | 1 | #144 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=frontiercode · How It Works · Data refreshed daily, snapshot 2026-10-20.