LongAudioSpan - Chain: leaderboard
LongAudioSpan chain score on long-form audio with acoustic anchors planted into it: a perception-to-reasoning chain of three multiple-choice links per item, credited only up to the first wrong link and averaged over the three duration tiers.
Metric: Chain score (%, mean over S/M/L tiers). Source: arxiv.org. Saturation forecast: Around 2029. 15 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3.5 Omni Plus | 29 | #111 |
| 2 | Gemini 2.5 Pro | 23.8 | #159 |
| 3 | Gemini 3.1 Pro (Preview) | 20.9 | #64 |
| 4 | Qwen3.5 Omni Flash | 20.2 | #225 |
| 5 | Gemini 3 Flash | 16.4 | #103 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=longaudiospan-chain · How It Works · Data refreshed daily, snapshot 2026-11-05.