Qwen-Audio-3.1-Realtime Report - Multilingual BBA - zh: leaderboard
Qwen-Audio-3.1-Realtime reportTable 3:14-language translated BigBenchAudio question evaluation, authorsown speech-to-speech accuracy, reportedpercent. English repeatsTable 2 BBA and is not duplicated. Mean is native average across 14 languages. GPTRealtime 2 effort unspecified and QwenAudio 3.0Plus/Flashvariant unspecified remain explicitreportscoped configurations. Metric: Accuracy (%)
Metric: Accuracy (%). Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 3 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen-Audio-3.1-Realtime Report Qwen-Audio-3.0-Realtime (Plus or Flash unspecified) | 90.9 | |
| 2 | Qwen-Audio-3.1-Realtime | 90.9 | |
| 3 | Qwen-Audio-3.1-Realtime Report GPT-Realtime-2 (effort unspecified) | 81.1 |
Interactive version: theaggregate.ai/benchmark?slug=qwen-audio-3-1-realtime-report-multilingual-bba-zh · How It Works · Data refreshed daily, snapshot 2026-11-05.