STEMVerse - Biology - Cell Biology: leaderboard
Metric: Few-shot accuracy (%). Source: arxiv.org. Saturation forecast: Estimated already saturated. 16 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen3-14B-Base | 53.08 |
| 2 | Qwen 2.5 14B Instruct | 49.62 |
| 3 | Qwen3-8B-Base | 49.62 |
| 4 | Qwen 2.5 14B | 46.15 |
| 5 | Qwen 2.5 7B | 44.62 |
| 6 | Qwen3-4B-Base | 42.69 |
| 7 | Llama 3.1 8B Instruct | 40.77 |
| 8 | Llama 3.1 8B | 38.85 |
| 9 | Qwen 2.5 3B | 35.38 |
| 10 | Qwen 2.5 3B Instruct | 34.23 |
| 11 | Qwen 2.5 7B Instruct | 27.69 |
| 12 | Llama 3.2 3B Instruct | 27.31 |
| 13 | Llama 3.2 3B | 23.85 |
Interactive version: theaggregate.ai/benchmark?slug=stemverse-biology-cell-biology · How It Works · Data refreshed daily, snapshot 2026-09-25.