LabBench (Gamow Labs): leaderboard
Metric: Mean share of binary rubric criteria passed (%) over 20 held-out wet-lab decision tasks (406 criteria), one run per task in each vendor's harness, judged by Codex with GPT-5.6 Sol. Source: gamowlabs.com. Saturation forecast: Around 2029. 5 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-6 | 40.8 |
| 2 | Claude Opus 5.5 (Max) | 40.5 |
| 3 | Grok 4.7 | 29.6 |
| 4 | Muse Spark 1.3 (High) | 28.2 |
| 5 | Gemini 3.8 Flash | 18.4 |
Interactive version: theaggregate.ai/benchmark?slug=labbench-gamow-labs · How It Works · Data refreshed daily, snapshot 2026-10-01.