FrontierCode 1.1
- OpenAI
- Anthropic
- Reported in a later post or card
Results
| Model | Score | Source | Source check |
|---|---|---|---|
| Claude Haiku 5.57 Oct 2026 · Anthropic | 58.4% | Claude Haiku 5.5 System Cardmax effort; run by Cognition in Claude Code; mean of 5 runs | Source checkedA person checked this number against its source on 20 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. |
| Claude Sonnet 5.528 Sep 2026 · Anthropic | 64.4% | Claude Sonnet 5.5 System Cardxhigh effort; run by Cognition in Claude Code | 1 other sourceThe same number is printed in 1 other lab document, linked below. |
| Claude Sonnet 5.528 Sep 2026 · Anthropic | 59.1% | Claude Sonnet 5.5 System Cardmax effort; run by Cognition in Claude Code; another report, not charted | 1 other sourceThe same number is printed in 1 other lab document, linked below. |
| Claude Opus 5.522 Sep 2026 · Anthropic | 65.3% | Claude Opus 5.5 System Cardmedium effort; run by Cognition in Claude Code | Source checkedA person checked this number against its source on 20 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. |
| Claude Opus 5.522 Sep 2026 · Anthropic | 63.6% | Claude Opus 5.5 System Cardmax effort; run by Cognition in Claude Code; another report, not charted | Source checkedA person checked this number against its source on 20 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. |
| GPT-6 Astra3 Sep 2026 · OpenAI | 64.5% | GPT-6 Astra: A new generation of intelligence | Source checkedA person checked this number against its source on 20 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. |
| Claude Fable 5.11 Sep 2026 · Anthropic | 63.6% | Claude Fable 5.1 and Claude Mythos 5.1 System Cardmedium effort | Source checkedA person checked this number against its source on 19 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. |
| Claude Opus 524 Jul 2026 · Anthropic | 63.6% | Claude Opus 5 System Cardrun and scored by Cognition; medium effort, the best of five effort levels; mean of 5 runs | 1 other sourceThe same number is printed in 1 other lab document, linked below. |
| GPT-5.6 Sol9 Jul 2026 · OpenAI | 60.6% | GPT-6 Astra: A new generation of intelligencereported later | Source checkedA person checked this number against its source on 20 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. |
Benchmark
cognition.comHigher is better
CognitionIntroduced 7 Jul 2026