Skip to content

Benchmark

Diarization · Macro cpWER

Scores are percentages. Lower is better.

Models measured
2
Scores recorded
2
Sources
1
IQ weight
No weight published

Score distribution

18.5%
Mean
18.6%
Median
0.1%
Std dev
18.4%
10th percentile
18.6%
90th percentile
18.4%
Min
18.4%18.6%
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
1Voxtral Small Transcribe 3Voxtral 3 official release chart · macro cpWER · AMI-IHM/SDM, Switchboard/CallHome, AISHELL-4· Reported by Mistral AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.18.4%
2Voxtral Mini Transcribe 3Voxtral 3 official release chart · macro cpWER · same public diarization set· Reported by Mistral AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.18.6%

Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links.