TraceAV-Bench - Information Retrieval: leaderboard

The Information Retrieval sub-task of TraceAV-Bench (2026), part of its audio-visual joint reasoning dimension: multiple-choice questions over long videos whose answers chain several timestamped evidence hops. The score is accuracy in percent with full audio and video input; higher is better.

Metric: Accuracy (%) on the 140 Information Retrieval (IR) questions, four-option multiple choice over long audio-visual videos (10 to 140 minutes), exact match of the selected option set, each model at its official inference setting, with the full audio and video input; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 15 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Gemini 2.5 Pro83.57#276
2Gemini 3.1 Pro (Preview)83.57#99
3Gemini 3 Flash82.14#175
4Gemini 2.5 Flash75#463
5Gemini 2.0 Flash66.43#723
6Ming-Flash-Omni-2.056.43#1266
7OmniVinci-9B49.29
8Qwen 3 Omni 30B A3B47.14#847
9Qwen2.5-Omni-7B46.43#1884
10MiniCPM-o-4.545.71#1252
11video-SALMONN 242.14#2687
12Gemma 4 E4B39.29#2184
13HumanOmni-7B37.86#3308
14Baichuan-Omni-1.537.14#3285
15VideoLLaMA2.1-7B-AV36.43#4204

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

Interactive version: theaggregate.ai/benchmark?slug=traceav-bench-information-retrieval · How It Works · Data refreshed daily, snapshot 2026-11-21.