reasoning benchmark
Big Bench Audio
Big Bench Audio is an audio reasoning benchmark adapted from a subset of Big Bench Hard, with text questions converted to spoken audio. It evaluates the reasoning ability of speech-to-speech and audio language models on tasks delivered as audio input, with accuracy scored by an independent evaluation (Artificial Analysis).
Updated Aug 7, 2026
Published models1
Registry coverage1
MetricScore
EvidenceB
Big Bench Audio leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | AM Nova 2 SonicAmazon | 87.0% | 100.0% | 1 | C |
Big Bench Audio highlights
The top published results on this benchmark's own scale.
What is Big Bench Audio?
Definition and scoring fields from the benchmark registry.
Big Bench Audio is an audio reasoning benchmark adapted from a subset of Big Bench Hard, with text questions converted to spoken audio. It evaluates the reasoning ability of speech-to-speech and audio language models on tasks delivered as audio input, with accuracy scored by an independent evaluation (Artificial Analysis).
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- Big Bench Audio
- Modality
- audio
- Primary category
- reasoning
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- big-bench-audio|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about Big Bench Audio.
Which model scores highest on Big Bench Audio?
Nova 2 Sonic is currently ranked first with 87.0%.
What does Big Bench Audio measure?
Big Bench Audio is an audio reasoning benchmark adapted from a subset of Big Bench Hard, with text questions converted to spoken audio. It evaluates the reasoning ability of speech-to-speech and audio language models on tasks delivered as audio input, with accuracy scored by an independent evaluation (Artificial Analysis).
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
1 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is preserved as source-native evidence but is not eligible for the current overall score.