OpenAIMistral AIAudio matchupSpeech to text
GPT-Live-Transcribevs.Voxtral Mini Transcribe Realtime
Which model is better?
GPT-Live-Transcribe is slightly ahead in independent tests, but the lead is not certain.
Pick GPT-Live-Transcribe for
Pick Voxtral Mini Transcribe Realtime for
Shared tests
Independent evaluators and leaderboards only, the newest shared snapshot per evaluator. With 95 % ranges a lead counts when they do not overlap; for hit rates, when it holds in a 95 % test. Without any uncertainty figure at most “slightly ahead”.
AirCaps A5Sv2 ASR benchmark
GPT-Live-Transcribe slightly aheadGPT-Live-Transcribe is clearly ahead in 0 of 6 subtests and slightly in 6, Voxtral Mini Transcribe Realtime in none.
A5Sv2 ASR benchmark, DiPCo evaluation and development meetings
33,679 tasks, lower is better, as of October 17, 2026
A5Sv2 ASR benchmark, AMI scenario-only evaluation meetings
32,928 tasks, lower is better, as of October 17, 2026
A5Sv2 ASR benchmark, pooled WER across four corpora
133,027 tasks, lower is better, as of October 17, 2026
Show all 6 subtests
| Subtest | GPT-Live-Transcribe | Voxtral Mini Transcribe Realtime | Verdict |
|---|---|---|---|
| A5Sv2 ASR benchmark, DiPCo evaluation and development meetings (lower is better) | 39.9 % | 75.5 % | GPT-Live-Transcribe slightly ahead |
| A5Sv2 ASR benchmark, AMI scenario-only evaluation meetings (lower is better) | 30.2 % | 48.7 % | GPT-Live-Transcribe slightly ahead |
| A5Sv2 ASR benchmark, pooled WER across four corpora (lower is better) | 34.7 % | 50.6 % | GPT-Live-Transcribe slightly ahead |
| A5Sv2 ASR benchmark, unweighted macro mean across four corpora (lower is better) | 34.7 % | 50.5 % | GPT-Live-Transcribe slightly ahead |
| A5Sv2 ASR benchmark, Mega-ASR fixed robustness selection (lower is better) | 27.8 % | 32.5 % | GPT-Live-Transcribe slightly ahead |
| A5Sv2 ASR benchmark, NOTSOFAR eval-small meetings (lower is better) | 40.8 % | 45.3 % | GPT-Live-Transcribe slightly ahead |
Source · latest October 17, 2026
Vals AI
GPT-Live-Transcribe slightly aheadGPT-Live-Transcribe is clearly ahead in 0 of 3 subtests and slightly in 3, Voxtral Mini Transcribe Realtime in none.
VoiceCodeBench · Overall
as of October 31, 2026
VoiceCodeBench · WER
lower is better, as of October 31, 2026
VoiceCodeBench · CTEM
as of October 31, 2026
Source · latest October 31, 2026
Artificial Analysis
GPT-Live-Transcribe slightly aheadArtificial Analysis AA-WER Streaming (final transcription)
lower is better, as of October 15, 2026
Source · latest October 15, 2026
All data side by side
Prices are verbatim from the provider, because audio models bill per minute, character or token.
| GPT-Live-Transcribe | Voxtral Mini Transcribe Realtime | |
|---|---|---|
| Task | Speech to text | Speech to text |
| Developer | OpenAI | Mistral AI |
| Status | Active | Active |
| Released | July 28, 2026 | n/a |
| Weights | proprietary | open |
| Runs locally | no | yes |
| Price | $0.017 per audio minute | n/a |
| Popularity (out of 100) | 39 | 18 |