Benchmark
Artificial Analysis Full Duplex Bench · Conversational Dynamics
Scores are percentages. Higher is better.
Benchmark sourceScores from BenchBoard docs.
- Models measured
- 1
- Scores recorded
- 1
- Sources
- 1
- IQ weight
- No weight published
Score distribution
- 97%
- Mean
- 97%
- Median
- 0%
- Std dev
- 97%
- 10th percentile
- 97%
- 90th percentile
- 97%
- Min
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Grok Voice 3.0October 13 release table · Artificial Analysis Full Duplex Bench subset· Reported by SpaceXAI, Artificial Analysis· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 97% |