STEMVerse - Physics - Fluid Mechanics: leaderboard
Metric: Few-shot accuracy (%). Source: arxiv.org. Saturation forecast: Estimated already saturated. 16 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen3-14B-Base | 32.46 |
| 2 | Qwen3-8B-Base | 23.98 |
| 3 | Qwen 2.5 14B | 20.76 |
| 4 | Llama 3.1 8B | 18.13 |
| 5 | Qwen 2.5 14B Instruct | 17.84 |
| 6 | Llama 3.1 8B Instruct | 17.54 |
| 7 | Qwen 2.5 7B | 16.67 |
| 8 | Llama 3.2 3B Instruct | 13.74 |
| 9 | Qwen 2.5 3B | 13.74 |
| 10 | Llama 3.2 3B | 13.45 |
| 11 | Qwen3-4B-Base | 11.99 |
| 12 | Qwen 2.5 7B Instruct | 11.7 |
| 13 | Qwen 2.5 3B Instruct | 11.44 |
Interactive version: theaggregate.ai/benchmark?slug=stemverse-physics-fluid-mechanics · How It Works · Data refreshed daily, snapshot 2026-09-25.