SYNCR: leaderboard
Metric: Average exact-match accuracy (%) over the eight SYNCR tasks, 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: Around February 2028. 19 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Flash | 52.5 |
| 2 | GPT-5.4 | 50.5 |
| 3 | Qwen 3 VL 32B Instruct | 47.3 |
| 4 | Qwen 3 VL 8B Instruct | 42.4 |
| 5 | Qwen 2.5 VL 72B Instruct | 40.6 |
| 6 | Qwen 2.5 VL 32B Instruct | 36.7 |
| 7 | Qwen 3 VL 4B Instruct | 35.6 |
| 8 | Qwen 2.5 VL 7B Instruct | 35.1 |
Interactive version: theaggregate.ai/benchmark?slug=syncr · How It Works · Data refreshed daily, snapshot 2026-10-07.