FrontierCode Diamond (Fable/Mythos): leaderboard
Hardest 50 FrontierCode production-code tasks from Cognition, measuring whether maintainers would merge model PRs using blocker criteria and quality rubrics. This slice reports Anthropic's own runs from the Claude Fable 5 / Mythos 5 system card; small frontier-model pool.
Metric: Score (%). Source: www-cdn.anthropic.com. Saturation forecast: Around May 2028. 4 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Fable 5 | 29.3 | #16 |
| 2 | Claude Mythos 5 | 29.3 | #2 |
| 3 | Claude Opus 4.8 | 13.4 | #37 |
| 4 | GPT-5.5 | 5.7 | #31 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=frontiercode-diamond-fable-mythos · How It Works · Data refreshed daily, snapshot 2026-10-20.