Epoch AI - Science Qa: leaderboard
Multiple-choice science benchmark across natural science topics. Originally multimodal (text + images/diagrams) but Epoch AI tracks only the text-only subset for LLMs.
Metric: Score. Source: epoch.ai. Saturation forecast: Estimated already saturated. 23 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4o (2024-05-13) | 88.5 | #325 |
| 2 | Median Human | 88.4 | |
| 3 | Gemini 1.0 Pro | 79.7 | #933 |
| 4 | text-davinci-001 | 74.04 | #1512 |
| 5 | Claude 3 Haiku (20240307) | 72 | #771 |
| 6 | Llama 2 13B | 55.78 | #1436 |
| 7 | LLaMA-13B | 43.33 | #1516 |
| 8 | Llama 2 7B | 43.08 | #1520 |
| 9 | LLaMA-7B | 36.19 | #1564 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=epoch-ai-science-qa · How It Works · Data refreshed daily, snapshot 2026-10-20.