AA FrontierCode 2.0: leaderboard
Artificial Analysis runs all 132 private FrontierCode 2.0 tasks in Stirrup, with three attempts per task and a 500-turn cap. Blocking rubric failures or solution-lookup flags score zero; other attempts receive weighted rubric credit from the GPT-6 Luna (medium) judge. The mean equally weights tasks. Scores are percentages; earlier FrontierCode protocols remain separate.
Metric: Guardrail-gated Rubric Score (%). Source: artificialanalysis.ai. Saturation forecast: Around September 2028. 78 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6 Astra (Max) | 42.37 | #5 (GPT-6 Astra) |
| 2 | GPT-6.1 Terra (Max) | 42.32 | #13 (GPT-6.1 Terra) |
| 3 | GPT-6 Astra (xHigh) | 41.4 | #5 (GPT-6 Astra) |
| 4 | GPT-6.1 Sol (Max) | 40.63 | #7 (GPT-6.1 Sol) |
| 5 | Gemini 4 Carbon (High) | 40.56 | #2 (Gemini 4 Carbon) |
| 6 | GPT-6 Astra (High) | 40.14 | #5 (GPT-6 Astra) |
| 7 | GPT-6.1 Sol (xHigh) | 38.52 | #7 (GPT-6.1 Sol) |
| 8 | Muse Wonder Full (Max) | 37.76 | #11 (Muse Wonder Full) |
| 9 | GPT-6.1 Terra (High) | 37.46 | #13 (GPT-6.1 Terra) |
| 10 | GPT-6.1 Terra (xHigh) | 36.86 | #13 (GPT-6.1 Terra) |
| 11 | GPT-6.1 Sol (High) | 36.74 | #7 (GPT-6.1 Sol) |
| 12 | GPT-6 Astra (Medium) | 36.58 | #5 (GPT-6 Astra) |
| 13 | Claude Haiku 5.5 (Max) | 34.93 | #40 (Claude Haiku 5.5) |
| 14 | GPT-6.1 Terra (Medium) | 34.54 | #13 (GPT-6.1 Terra) |
| 15 | GPT-6.1 Sol (Medium) | 33.39 | #7 (GPT-6.1 Sol) |
No result here: #3 Claude Opus 5.5, #10 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=aa-frontiercode-2-0 · How It Works · Data refreshed daily, snapshot 2026-10-29.