OpenAI GPT-5 System Card - Virology Capabilities Test — leaderboard
Metric: Mean Accuracy (%). Source: cdn.openai.com. 3 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | O3 (High) | 42 |
| 2 | GPT-5 (Thinking) | 41.4 |
| 3 | Human Expert | 22.6 |
Interactive version: theaggregate.ai/benchmark?slug=openai-gpt-5-system-card-virology-capabilities-test · How the rankings work · Data refreshed daily, snapshot 2026-07-22.