Skip to content

Benchmark

Big Bench Audio · Intelligence

Scores are percentages. Higher is better.

Models and systems measured
4
Scores recorded
4
Sources
1
IQ weight
No weight published

Score distribution

93.6%
Mean
97.7%
Median
7%
Std dev
81.4%
10th percentile
98.5%
90th percentile
81.4%
Min
81.4%98.5%
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#Model or systemScore
1Grok Voice 3.0October 13 release table · Artificial Analysis speech reasoning · 1,000 audio questions· Reported by SpaceXAI, Artificial Analysis· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.98.5%
2Gemini 3.8 Live Extended Thinkinghigh thinking · Gemini API · pass@1 · Google release prose· Reported by Google· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.97.7%
3GPT Realtime 2OpenAI release chart · high reasoning· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.96.6%

Not ranked

The source marks these results as descriptive, not capability scores, so we list them by name instead of ranking them.

Model or systemScore
GPT-Realtime-1.5OpenAI release chart · reasoning setting unspecified· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.81.4%

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.