LLM Stats (BioMysteryBench): leaderboard
LLM Stats mirror of BioMysteryBench, which works models through hard molecular biology problems and reports both the difficult subset and the subset human experts managed to solve.
Metric: Score (%). Source: llm-stats.com. Saturation forecast: Around January 2027. 7 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 5 | 90.1 | #12 |
| 2 | Claude Opus 5.5 | 89.3 | #3 |
| 3 | Claude Sonnet 5.5 | 89.2 | #11 |
| 4 | Hy4 preview | 71.3 | #38 |
| 5 | Gemini 3.8 Flash | 56.5 | #21 |
| 6 | Claude Fable 5 | 46.1 | #16 |
| 7 | Gemini 3.7 Flash | 43.5 | #23 |
No result here: #6 GPT-6 Astra, #7 GPT-6.1 Sol, #8 Gemini 4 Argon.
Interactive version: theaggregate.ai/benchmark?slug=llm-stats-biomysterybench · How It Works · Data refreshed daily, snapshot 2026-10-20.