Benchmarks · coding
FrontierCode 1.1 Extended
Mergeability of coding-agent changes on private maintainer-authored tasks, graded by unit tests, rubrics and verifiers (Cognition).
45 models · 134 results · data as of 2026-10-29 · cognition.com ↗
Leaderboard
Opus 5.5 medium
More from this benchmark
2 published measuresTime per task24 results · lower is better
Output tokens per task134 results · lower is better
All results
134 rows| # | Model | Effort | Score | Subscription $ / task | API $ / task | Time | Output tokens | Harness |
|---|---|---|---|---|---|---|---|---|
| 1 | Opus 5.5 | medium | 65.3% | $0.010 | $0.67 | 5.6 min | 15k | claude-code |
| 2 | Opus 5.5 | high | 65.2% | $0.014 | $0.90 | 6.3 min | 21k | claude-code |
| 3 | Fable 5 | xhigh | 64.9% | $1.19 | $11 | 48k | claude-code | |
| 4 | GPT-6 Astra | max | 64.5% | $0.34 | $3.93 | 26k | codex | |
| 5 | Sonnet 5.5 | xhigh | 64.4% | $0.019 | $1.24 | 10 min | 38k | claude-code |
| 6 | Fable 5 | high | 64.3% | $0.87 | $7.71 | 36k | claude-code | |
| 7 | Fable 5 | max | 63.6% | $1.74 | $15 | 68k | claude-code | |
| 8 | Opus 5 | medium | 63.6% | $0.061 | $3.51 | 28k | claude-code | |
| 9 | Fable 5.1 | medium | 63.6% | $0.42 | $2.68 | 21k | claude-code | |
| 10 | Opus 5.5 | max | 63.6% | $0.080 | $5.28 | 22 min | 144k | claude-code |
| 11 | Opus 5.5 | xhigh | 63.5% | $0.028 | $1.86 | 9.5 min | 49k | claude-code |
| 12 | GPT-6 Astra | high | 63.1% | $0.22 | $2.62 | 12k | codex | |
| 13 | Fable 5 | medium | 62.8% | $0.66 | $5.84 | 28k | claude-code | |
| 14 | Fable 5.1 | high | 62.7% | $0.67 | $4.28 | 33k | claude-code | |
| 15 | SWE 2 | max | 62.5% | $0.94 | $0.94 | 60k | devin | |
| 16 | GPT-6 Astra | xhigh | 62.1% | $0.24 | $2.86 | 15k | codex | |
| 17 | Fable 5.1 | max | 62.0% | $1.67 | $11 | 78k | claude-code | |
| 18 | Fable 5.1 | low | 61.6% | $0.30 | $1.94 | 15k | claude-code | |
| 19 | Sonnet 5.5 | high | 61.5% | $0.005 | $0.35 | 5.6 min | 11k | claude-code |
| 20 | Fable 5.1 | xhigh | 61.4% | $1.20 | $7.70 | 57k | claude-code | |
| 21 | Grok 4.6 | high | 61.3% | $0.25 | $2.38 | 32k | grok-build | |
| 22 | Fable 5 | low | 60.8% | $0.45 | $4.03 | 20k | claude-code | |
| 23 | GPT-6 Sol | max | 60.7% | $0.10 | $1.66 | 34k | codex | |
| 24 | Gemini 4 Argon | low | 60.6% | $2.19 | $6.14 | 21 min | 43k | devin-cli |
| 25 | GPT-5.6 Sol | max | 60.6% | $0.23 | $4.25 | 28k | codex | |
| 26 | Gemini 4 Argon | medium | 60.5% | $2.00 | $5.62 | 21 min | 43k | devin-cli |
| 27 | GPT-6.1 Sol | medium | 60.4% | $0.036 | $0.31 | 13 min | 9.5k | codex |
| 28 | GPT-6.1 Sol | xhigh | 60.4% | $0.057 | $0.49 | 15 min | 16k | codex |
| 29 | Opus 5.5 | low | 60.3% | $0.005 | $0.35 | 4.7 min | 7.6k | claude-code |
| 30 | GPT-6 Astra | medium | 60.3% | $0.18 | $2.10 | 9.0k | codex |