STEMVerse - Physics - Bloom Level: Understand Leaderboard · The Aggregate

STEMVerse - Physics - Bloom Level: Understand: leaderboard

Metric: Few-shot accuracy (%). Source: arxiv.org. Saturation forecast: Estimated already saturated. 16 models tracked.

Top models

#	Model	Score
1	Qwen3-14B-Base	86.67
2	Qwen3-8B-Base	83.33
3	Llama 3.1 8B	80
4	Qwen 2.5 14B Instruct	80
5	Qwen 2.5 7B Instruct	76.67
6	Llama 3.1 8B Instruct	73.33
7	Qwen 2.5 14B	73.33
8	Qwen 2.5 7B	66.67
9	Qwen 2.5 3B	63.33
10	Llama 3.2 3B Instruct	56.67
11	Qwen 2.5 3B Instruct	53.33
12	Llama 3.2 3B	43.33
13	Qwen3-4B-Base	10

Interactive version: theaggregate.ai/benchmark?slug=stemverse-physics-bloom-level-understand · How It Works · Data refreshed daily, snapshot 2026-09-25.