Biomni-Bench (Humanlaya): leaderboard
Humanlaya evaluation of frontier models on Biomni biomedical-agent data-analysis tasks, each model in its native agent harness (Claude Code / Codex / Gemini CLI); mean score 0-100 over 3 runs.
Metric: Score (0-100). Source: humanlaya.com. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.7 | 73.34 | #51 |
| 2 | Claude Opus 4.6 | 69.51 | #67 |
| 3 | GPT-5.4 | 68.69 | #83 |
| 4 | GPT-5.5 | 67.59 | #33 |
| 5 | Gemini 3.1 Pro (Preview) | 49.83 | #64 |
Interactive version: theaggregate.ai/benchmark?slug=biomni-bench-humanlaya · How It Works · Data refreshed daily, snapshot 2026-11-05.