BloomBench - English: leaderboard

Metric: Micro accuracy (%) on all 7,747 English BloomBench items across the six Bloom levels, four-option image-question multiple choice, zero-shot, temperature 0, the chosen letter parsed from the generated answer (Regex-based Answer Extraction; unparseable answers count as wrong); chance 25; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 8 models tracked.

Top models

#ModelScore
1Gemma 4 31B89.8
2Gemma 3 27B88.3
3Gemma 4 26B A4B87.6
4Qwen 2.5 VL 7B86.9
5Gemma 3 12B86.7
6Qwen 2 VL 7B85.4
7GPT-4o Mini82.4
8Gemma 3 4B79.6

Interactive version: theaggregate.ai/benchmark?slug=bloombench-english · How It Works · Data refreshed daily, snapshot 2026-09-29.