GenderBench - GEST Creative Gender Imbalance: leaderboard

In the GEST creative-writing task, the number is how far the share of male protagonists is from 0.5, regardless of stereotype. A balanced writer scores 0.

Metric: Distance of the male-protagonist rate from 0.5 (GEST traits) (0-1 normalized, lower is better). Source: genderbench.readthedocs.io. 12 models tracked.

Top models

#ModelScoreOverall rank
1Gemini 2.0 Flash Lite0.07#1032
2Mistral 7B Instruct (v0.3)0.1#6653
3Phi-40.14#1913
4GPT-4o0.17#741
5Gemma 2 27B (IT)0.22#1782
6Gemini 2.0 Flash0.26#704
7Llama 3.1 8B Instruct0.26#4246
8Gemma 2 9B (IT)0.26#2262
9Mistral Small 30.27#1777
10GPT-4o Mini0.29#1623
11Llama 3.3 70B Instruct0.34#1357
12Claude 3.5 Haiku0.4#1553

Interactive version: theaggregate.ai/benchmark?slug=genderbench-gest-creative-gender-imbalance · How It Works · Data refreshed daily, snapshot 2026-11-13.