Voxtral 3 Release - Earnings22 Keyphrase Recall - Without Context: leaderboard
Voxtral 3 release evaluation (2026-10-20): exact chart marks retain model, series and score. FLEURS 45 macro-averages WER, using CER for Japanese, Korean, Thai and Chinese; language forced/inferred runs stay separate. Diarization averages cpWER over AMI-IHM, AMI-SDM, Switchboard, CallHome and AISHELL-4 with 30-minute clips. Context-bias recall uses Earnings22. Realtime latency configurations remain distinct. TTS tests retain their release-specific populations and judge; chart x-axis prices are never scores. Recall (%).
Metric: Recall (%). Source: mistral.ai. Saturation forecast: Rough model projection: around 2026. 3 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Scribe v2 | 84 | |
| 2 | Voxtral Small Transcribe 3 | 81.4 | |
| 3 | Gemini 3.5 Transcribe | 80.9 |
Interactive version: theaggregate.ai/benchmark?slug=voxtral-3-release-earnings22-keyphrase-recall-without-context · How It Works · Data refreshed daily, snapshot 2026-10-21.