Skip to content

Benchmark

Artificial Analysis Speech Agent Arena · Elo

Scores are Elo ratings. Higher ratings are better.

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

Score distribution

1,067
Mean
1,070
Median
3
Std dev
1,064
10th percentile
1,070
90th percentile
1,064
Min
1,0641,070
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
1Grok Voice 3.0October 13 release table · Speech Agent Arena snapshot· Reported by SpaceXAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,070
2Grok Voice 3.0Artificial Analysis live leaderboard · October 24 snapshot· Reported by Artificial Analysis· via BenchBoard docs1,064

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.