TraceAV-Bench - Spatial Reasoning: leaderboard

The Spatial Reasoning sub-task of TraceAV-Bench (2026), part of its visual-centric 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 165 Spatial Reasoning (SR) 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 3.1 Pro (Preview)73.94#99
2Gemini 3 Flash73.33#175
3Gemini 2.5 Pro68.48#276
4Gemini 2.5 Flash66.06#463
5Ming-Flash-Omni-2.065.45#1266
6MiniCPM-o-4.560.61#1252
7Qwen 3 Omni 30B A3B58.18#847
8Gemma 4 E4B55.15#2184
9HumanOmni-7B55.15#3308
10OmniVinci-9B55.15
11Gemini 2.0 Flash54.55#723
12Qwen2.5-Omni-7B49.7#1884
13video-SALMONN 248.48#2687
14VideoLLaMA2.1-7B-AV38.79#4204
15Baichuan-Omni-1.537.58#3285

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-spatial-reasoning · How It Works · Data refreshed daily, snapshot 2026-11-21.