PolySpeech-100 - Low-Resource Languages: leaderboard
Metric: Accuracy (%; spoken four-option reading comprehension from Belebele: the audio carries the passage, question and options; text instruction as system prompt; mean over 81 long-tail languages). Source: arxiv.org. Saturation forecast: Around December 2026. 9 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Flash (Preview) | 84.61 |
| 2 | MiniCPM-o-4.5 | 45.12 |
| 3 | Qwen2.5-Omni-7B | 40.18 |
Interactive version: theaggregate.ai/benchmark?slug=polyspeech-100-low-resource-languages · How It Works · Data refreshed daily, snapshot 2026-09-26.