FMSU-Bench - Gender (English): leaderboard

FMSU-Bench (2026) tests fine-grained speech understanding in Chinese and English with multiple-choice questions whose distractors include subtle acoustic confusions and text-only semantic traps, plus a tagged transcription task. This board is accuracy on identifying the gender of the speaker in the English subset; percent, higher is better.

Metric: Accuracy (%) of identifying the gender of the speaker (GEN), on the FMSU-Bench English subset (500 to 1,000 manually verified items per task, from audiovisual sources), multiple-choice questions whose distractors include fine-grained acoustic confusions and text-only semantic traps; higher is better. Source: arxiv.org. Saturation forecast: Not forecast. 10 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Qwen 3 Omni 30B A3B99.5#851
2Qwen2.5-Omni-7B99.3#1888
3Audio Flamingo 397.6
4MiMo-Audio96.9#934
5Kimi-Audio95.6#1722
6Gemini 2.5 Flash93.6#466
7Gemini 3 Flash88.3#180
8FMSU-Bench Step-Audio 2 (checkpoint unspecified)87.5
9Gemini 3.1 Pro (Preview)85.3#102
10Qwen2-Audio (FMSU-Bench checkpoint unspecified)36.8

No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=fmsu-bench-gender-english · How It Works · Data refreshed daily, snapshot 2026-11-22.