Benchmarks · coding
FrontierCode 2.0-AA
AA's run of Cognition's FrontierCode 2.0 (132 private tasks in real open-source repositories, three attempts each): would a maintainer merge the change? An attempt failing any blocking check scores 0. Separate from Cognition's own leaderboard, which uses different harnesses.
22 models · 61 results · data as of 2026-10-29 · artificialanalysis.ai ↗
Leaderboard
Opus 5.5 max
More from this benchmark
4 published measuresScore vs any published measure4 measures: sub-scores, splits, counts
Breakdown
darker is better within each column · click a column to sort| Model | ||||
|---|---|---|---|---|
| Opus 5.5 max, fallback | 4,566,850,091 | 7,912,780 | 18,486,777 | 4,512,172,119 |
| GPT-6 Astra max | 390,224,479 | 2,034,335 | 2,861,315 | 372,853,166 |
| GPT-6.1 Terra max | 1,586,763,300 | 3,635,198 | 5,359,529 | 1,558,480,336 |
| Sonnet 5.5 max, fallback | 7,515,883,580 | 11,727,454 | 29,982,620 | 7,437,592,141 |
| GPT-6.1 Sol max | 449,271,572 | 2,089,214 | 3,157,401 | 430,526,708 |
| Gemini 4 Carbon high | 2,645,830,649 | 4,350,614 | 12,514,814 | 2,595,119,244 |
| Fable 5.1 xhigh, fallback | 2,088,852,514 | 5,975,268 | 6,639,021 | 2,057,970,781 |
| Haiku 5.5 max | 3,425,414,487 | 7,643,360 | 26,007,861 | 3,372,943,041 |
| Grok 4.7 xhigh | 2,542,145,904 | 4,096,174 | 11,493,398 | 2,507,165,153 |
| DeepSeek V4.1 Flash 1017 max | 3,390,278,953 | 5,530,732 | 9,794,666 | 3,358,168,087 |
| Gemini 4 Argon high | 3,052,442,515 | 3,447,687 | 17,189,859 | 2,993,928,314 |
| GPT-6 Luna max | 1,449,282,549 | 2,231,042 | 6,987,768 | 1,422,863,628 |
| MiMo V2.6 Pro | 1,514,317,971 | 2,900,711 | 12,471,686 | 1,489,238,665 |
| Muse Spark 1.3 max | 1,998,878,243 | 4,065,842 | 6,039,305 | 1,971,173,983 |
| MiniMax M3 | 2,108,051,866 | 3,417,340 | 10,207,771 | 2,078,319,579 |
| MiMo V2.6 Flash | 1,672,387,758 | 3,058,630 | 12,673,910 | 1,644,933,600 |
| GLM 5.3 max | 2,683,783,478 | 4,363,236 | 9,615,126 | 2,656,208,250 |
| Gemini 3.8 Flash high | 3,674,659,215 | 3,051,964 | 12,522,288 | 3,634,279,278 |
| Qwen 3.8 27B xhigh | 2,711,374,329 | 5,186,171 | 9,363,394 | 2,683,545,178 |
| Gemini 3.5 Flash-Lite high | 2,198,950,750 | 2,567,356 | 5,363,567 | 2,167,326,459 |
| Inkling Small | 820,300,258 | 1,574,398 | 3,527,674 | 808,931,942 |
| Inkling xhigh | 2,095,144,128 | 2,159,204 | 5,698,266 | 2,068,491,014 |
All results
61 rows| # | Model | Effort | FrontierCode 2.0-AA | Subscription $ / task | API $ / task | Harness |
|---|---|---|---|---|---|---|
| 1 | Opus 5.5 | max | 46.4% | $0.20 | $13 | Stirrup |
| 2 | Opus 5.5 | xhigh | 43.8% | $0.098 | $6.49 | Stirrup |
| 3 | GPT-6 Astra | max | 42.4% | $0.56 | $6.54 | Stirrup |
| 4 | GPT-6.1 Terra | max | 42.3% | $1.29 | $1.29 | Stirrup |
| 5 | Sonnet 5.5 | max | 41.7% | $0.17 | $12 | Stirrup |
| 6 | GPT-6 Astra | xhigh | 41.4% | $0.35 | $4.04 | Stirrup |
| 7 | GPT-6.1 Sol | max | 40.6% | $0.13 | $1.13 | Stirrup |
| 8 | Gemini 4 Carbon | high | 40.6% | $12 | $12 | Stirrup |
| 9 | GPT-6 Astra | high | 40.1% | $0.32 | $3.77 | Stirrup |
| 10 | GPT-6.1 Sol | xhigh | 38.5% | $0.082 | $0.70 | Stirrup |
| 11 | GPT-6.1 Terra | high | 37.5% | $0.39 | $0.39 | Stirrup |
| 12 | GPT-6.1 Terra | xhigh | 36.9% | $0.57 | $0.57 | Stirrup |
| 13 | GPT-6.1 Sol | high | 36.7% | $0.074 | $0.63 | Stirrup |
| 14 | GPT-6 Astra | medium | 36.6% | $0.24 | $2.80 | Stirrup |
| 15 | Opus 5.5 | high | 35.9% | $0.038 | $2.49 | Stirrup |
| 16 | Fable 5.1 | xhigh | 35.7% | $1.97 | $13 | Stirrup |
| 17 | Sonnet 5.5 | xhigh | 35.6% | $0.041 | $2.71 | Stirrup |
| 18 | Haiku 5.5 | max | 34.9% | $0.010 | $0.46 | Stirrup |
| 19 | GPT-6.1 Terra | medium | 34.5% | $0.26 | $0.26 | Stirrup |
| 20 | GPT-6.1 Sol | medium | 33.4% | $0.048 | $0.41 | Stirrup |
| 21 | GPT-6.1 Terra | none | 33.4% | $0.43 | $0.43 | Stirrup |
| 22 | Opus 5.5 | medium | 33.1% | $0.026 | $1.71 | Stirrup |
| 23 | Fable 5.1 | max | 31.6% | $2.54 | $16 | Stirrup |
| 24 | Fable 5.1 | high | 30.3% | $1.16 | $7.43 | Stirrup |
| 25 | Haiku 5.5 | xhigh | 29.9% | $0.005 | $0.26 | Stirrup |
| 26 | Fable 5.1 | medium | 28.8% | $0.75 | $4.78 | Stirrup |
| 27 | Grok 4.7 | xhigh | 28.7% | $1.28 | $14 | Stirrup |
| 28 | DeepSeek V4.1 Flash 1017 | max | 28.5% | free | $0.52 | Stirrup |
| 29 | Grok 4.7 | high | 27.8% | $1.03 | $11 | Stirrup |
| 30 | GPT-6 Astra | low | 26.2% | $0.11 | $1.34 | Stirrup |