TraceAV-Bench - Cross-Modality Matching: leaderboard

The Cross-Modality Matching 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 85 Cross-Modality Matching (CMM) 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 2029. 15 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Gemini 2.5 Pro54.12#280
2Gemini 3.1 Pro (Preview)49.41#102
3Gemini 3 Flash49.41#180
4Ming-Flash-Omni-2.044.71#1258
5Gemini 2.0 Flash41.18#731
6MiniCPM-o-4.541.18#1251
7Gemini 2.5 Flash40#466
8Qwen 3 Omni 30B A3B40#851
9Qwen2.5-Omni-7B37.65#1888
10Gemma 4 E4B36.47#2199
11OmniVinci-9B35.29
12video-SALMONN 232.94#2695
13HumanOmni-7B25.88#3241
14VideoLLaMA2.1-7B-AV25.88#4199
15Baichuan-Omni-1.520#3189

No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=traceav-bench-cross-modality-matching · How It Works · Data refreshed daily, snapshot 2026-11-22.