Skip to content

Benchmark

MIR-1K Vocal-only · WER/CER

Scores are percentages. Lower is better.

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

Score distribution

4.8%
Mean
4.9%
Median
0.19%
Std dev
4.6%
10th percentile
4.9%
90th percentile
4.6%
Min
4.6%4.9%
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
1Qwen3.5 Omni PlusQwen3.5-Omni release · ASR WER/CER · lower is better· Reported by Qwen, Qwen Team· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.4.6%
2Qwen3.5 Omni FlashQwen3.5-Omni release · ASR WER/CER · lower is better· Reported by Qwen, Qwen Team· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.4.9%

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.