SYNCR - Sequential Ordering: leaderboard
Metric: Exact-match accuracy (%) on the SYNCR Sequential Ordering task (putting clips of one event in chronological order), zero-shot four-option multiple choice over 2 to 4 labelled input videos, exact match, deterministic decoding, on the paper's balanced subset of 200 questions per task; chance is 25; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 19 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Flash | 100 |
| 2 | Qwen 3 VL 32B Instruct | 93.5 |
| 3 | Qwen 3 VL 8B Instruct | 92.5 |
| 4 | Qwen 2.5 VL 72B Instruct | 75 |
| 5 | Qwen 3 VL 4B Instruct | 61.5 |
| 6 | GPT-5.4 | 60 |
| 7 | Qwen 2.5 VL 7B Instruct | 46 |
| 8 | Qwen 2.5 VL 32B Instruct | 36.5 |
Interactive version: theaggregate.ai/benchmark?slug=syncr-sequential-ordering · How It Works · Data refreshed daily, snapshot 2026-10-07.