Omni-DeepSearch - Video Search: leaderboard

Omni-DeepSearch (2026) tests audio-driven deep search: from one or more audio clips alone, a model must infer the audio entity, call text, image and video search tools and give a short verifiable answer, judged by a majority of three LLM judges. This board is accuracy on the 160 questions that require retrieving and reasoning over a video, in percent; higher is better.

Metric: Accuracy (%) on the 160 single-audio video search questions (the model must retrieve and reason over a video), 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 February 2027. 12 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Gemini 3 Pro36.88#145
2Gemini 2.5 Pro20.62#276
3Gemini 3 Flash11.88#175
4Qwen 3.5 Omni Plus10.62#254
5MiMo-V2-Omni8.75#465
6MiMo-V2.58.13#383
7Gemini 2.5 Flash Lite3.13#964
8Qwen3.5 Omni Flash3.13#602
9Qwen3 Omni 30B A3B Instruct2.5#930
10Qwen2.5-Omni-7B1.88#1884
11Qwen2.5-Omni-3B0#3519
12Qwen 3 Omni 30B A3B (Thinking)0

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=omni-deepsearch-video-search · How It Works · Data refreshed daily, snapshot 2026-11-21.