STEMVerse - Biology - Biophysics and Biochemistry: leaderboard
Metric: Few-shot accuracy (%). Source: arxiv.org. Saturation forecast: Estimated already saturated. 16 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen3-14B-Base | 52.35 |
| 2 | Qwen3-4B-Base | 46.64 |
| 3 | Qwen3-8B-Base | 45.3 |
| 4 | Qwen 2.5 14B Instruct | 43.96 |
| 5 | Qwen 2.5 14B | 39.6 |
| 6 | Qwen 2.5 7B | 38.26 |
| 7 | Llama 3.1 8B Instruct | 35.91 |
| 8 | Llama 3.1 8B | 33.56 |
| 9 | Qwen 2.5 3B | 29.53 |
| 10 | Qwen 2.5 3B Instruct | 28.86 |
| 11 | Qwen 2.5 7B Instruct | 22.48 |
| 12 | Llama 3.2 3B | 21.14 |
| 13 | Llama 3.2 3B Instruct | 19.8 |
Interactive version: theaggregate.ai/benchmark?slug=stemverse-biology-biophysics-and-biochemistry · How It Works · Data refreshed daily, snapshot 2026-09-25.