MORPHOGEN - Hindi Corpus Gender Accuracy: leaderboard

Gender-aware morphological generation in Hindi on MORPHOGEN (2026): the model rewrites a first-person sentence for a speaker of the opposite grammatical gender while keeping its meaning, scored against human-corrected references. Corpus-level share of reference gendered terms generated correctly, in percent; higher is better.

Metric: Corpus-level gender accuracy (%): correctly generated gendered terms over all reference gendered terms of the test set, in the GENFORM task of MORPHOGEN: rewrite a first-person Hindi sentence for a speaker of the opposite grammatical gender while keeping its meaning (both directions, human-corrected references), zero-shot with the model chat template, temperature 0.1; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 15 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Phi-495.1#1931
2GPT-4o Mini93.36#1667
3Llama 3.3 70B91.4#1514
4Qwen 3 32B90.38#972
5Llama 3.1 8B89.21#5196
6Qwen 3 8B87.82#1873
7Qwen 3 14B85.8#1269
8Gemma 2 9B84.39#3201
9Gemma 3 27B82.56#1663
10Gemma 3 12B80.99#2059
11Qwen 3 4B68.51#2348
12Gemma 2 2B65.41#8415
13Gemma 3 4B64.58#5339
14Llama 3.2 3B49.72#7877
15Qwen 2.5 0.5B0.21#9144

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=morphogen-hindi-corpus-gender-accuracy · How It Works · Data refreshed daily, snapshot 2026-11-21.