TraceAV-Bench - Spatiotemporal Localization: leaderboard

The Spatiotemporal Localization 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 227 Spatiotemporal Localization (SL) 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 December 2027. 15 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Gemini 3.1 Pro (Preview)51.54#99
2Gemini 2.5 Pro48.02#276
3Qwen2.5-Omni-7B37.44#1884
4MiniCPM-o-4.537.44#1252
5HumanOmni-7B35.68#3308
6OmniVinci-9B33.48
7Qwen 3 Omni 30B A3B32.6#847
8Ming-Flash-Omni-2.031.28#1266
9video-SALMONN 231.28#2687
10Baichuan-Omni-1.530.4#3285
11Gemini 2.5 Flash29.07#463
12Gemini 3 Flash29.07#175
13Gemini 2.0 Flash25.11#723
14VideoLLaMA2.1-7B-AV22.91#4204
15Gemma 4 E4B16.74#2184

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-spatiotemporal-localization · How It Works · Data refreshed daily, snapshot 2026-11-21.