Skip to content

Benchmark

Voice Clone Similarity · In-house Multilingual

Higher scores are better.

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

Score distribution

0.8
Mean
0.8
Median
0
Std dev
0.8
10th percentile
0.8
90th percentile
0.8
Min

Leaderboard

#ModelScore
1Qwen3.5 Omni PlusQwen3.5-Omni release · voice clone similarity · cosine similarity · 9-language in-house set· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.0.8

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.