FMSU-Bench: 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. The overall score averages the 14 tasks over both languages, 13 accuracies and one transcription score; higher is better.

Metric: Average score (%) over the 14 tasks in both the Chinese and English subsets (13 multiple-choice accuracies and the PATA transcription score), over the 20,000-plus bilingual FMSU-Bench items; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 10 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Gemini 3.1 Pro (Preview)74#99
2Gemini 3 Flash71.9#175
3Qwen 3 Omni 30B A3B69.4#847
4Gemini 2.5 Flash69#463
5MiMo-Audio64.1#931
6Qwen2.5-Omni-7B59.7#1884
7Kimi-Audio54.3#1735
8FMSU-Bench Step-Audio 2 (checkpoint unspecified)48.7
9Audio Flamingo 347.6
10Qwen2-Audio (FMSU-Bench checkpoint unspecified)23.6

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

Interactive version: theaggregate.ai/benchmark?slug=fmsu-bench · How It Works · Data refreshed daily, snapshot 2026-11-21.