GlobeAudio - Chinese: leaderboard
Metric: Chinese (1,072 questions) accuracy (%) on GlobeAudio's naturalistic-audio multiple-choice questions written by native speakers; audio fed to the model directly, source-language prompts, predicted option label against the key. Source: arxiv.org. Saturation forecast: Around June 2027. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 70.34 |
| 2 | MiMo-V2.5 | 57.6 |
| 3 | Gemini 2.0 Flash | 55.13 |
| 4 | Voxtral-Small-24B-2507 | 51.77 |
| 5 | GPT-4o Audio | 50.75 |
| 6 | MERaLiON-2-10B | 48.79 |
| 7 | Gemma 3n E4B (IT) | 45.62 |
| 8 | Qwen2-Audio-7B-Instruct | 40.95 |
Interactive version: theaggregate.ai/benchmark?slug=globeaudio-chinese · How It Works · Data refreshed daily, snapshot 2026-09-29.