BioMysteryBench Human Difficult (Mythos 5.1 System Card): leaderboard
Biological reasoning and computational analysis on raw datasets, on problems unsolved by human experts but with objective ground truth. September 2026 configuration restricts package installation and network domains.
Metric: Score (%). Source: www-cdn.anthropic.com. Saturation forecast: Around 2028. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 5 | 51.8 | #16 |
| 2 | Claude Mythos 5 | 44.7 | #8 |
| 3 | Claude Mythos 5.1 | 44.1 | #2 |
| 4 | Claude Sonnet 5 | 39.4 | #65 |
| 5 | Gemini 3.1 Pro (Preview) | 32.9 | #64 |
| 6 | GPT-5.6 Sol | 28.8 | #22 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=biomysterybench-human-difficult-mythos-5-1-system-card · How It Works · Data refreshed daily, snapshot 2026-11-05.