TraceAV-Bench - Visual Counting: leaderboard
The Visual Counting 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 226 Visual Counting (VC) 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 2033. 15 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 41.15 | #99 |
| 2 | Gemini 2.5 Pro | 39.38 | #276 |
| 3 | Gemini 2.5 Flash | 39.38 | #463 |
| 4 | Ming-Flash-Omni-2.0 | 39.38 | #1266 |
| 5 | video-SALMONN 2 | 39.38 | #2687 |
| 6 | VideoLLaMA2.1-7B-AV | 38.94 | #4204 |
| 7 | MiniCPM-o-4.5 | 38.5 | #1252 |
| 8 | Qwen 3 Omni 30B A3B | 38.5 | #847 |
| 9 | Gemini 3 Flash | 36.73 | #175 |
| 10 | Qwen2.5-Omni-7B | 35.4 | #1884 |
| 11 | HumanOmni-7B | 34.96 | #3308 |
| 12 | OmniVinci-9B | 34.51 | |
| 13 | Gemma 4 E4B | 34.07 | #2184 |
| 14 | Gemini 2.0 Flash | 27.43 | #723 |
| 15 | Baichuan-Omni-1.5 | 26.99 | #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-visual-counting · How It Works · Data refreshed daily, snapshot 2026-11-21.