Omni-DeepSearch - Multi-Audio Text Search: leaderboard
Metric: Accuracy (%) on the 160 multi-audio text search questions (two to four clips sharing a mediator entity), audio-driven deep search: starting from audio clips alone, the model calls text, image and video search tools (at most 10 text or image searches and 3 video searches per question) and gives a short verifiable answer, judged by majority vote of GPT-5.4, Gemini-3-pro and Claude-Sonnet-4.6; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro | 40.63 |
| 2 | Gemini 3 Flash | 21.88 |
| 3 | Gemini 2.5 Pro | 13.75 |
| 4 | MiMo-V2.5 | 9.38 |
| 5 | Qwen 3.5 Omni Plus | 9.38 |
| 6 | Gemini 2.5 Flash Lite | 4.38 |
| 7 | MiMo-V2-Omni | 3.75 |
| 8 | Qwen3 Omni 30B A3B Instruct | 2.5 |
| 9 | Qwen3.5 Omni Flash | 2.5 |
| 10 | Qwen2.5-Omni-7B | 0 |
Interactive version: theaggregate.ai/benchmark?slug=omni-deepsearch-multi-audio-text-search · How It Works · Data refreshed daily, snapshot 2026-10-07.