SYNCR: leaderboard
SYNCR (2026) tests cross-video reasoning in multimodal models with synthetic videos from Habitat, Kubric and CLEVRER: temporal alignment, spatial tracking, comparative reasoning and holistic synthesis across 2 to 4 videos. The score is the mean exact-match accuracy (%) over its eight tasks, on 200 four-option questions per task; higher is better.
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 July 2027. 18 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Flash | 52.5 | #175 |
| 2 | GPT-5.4 | 50.5 | #135 |
| 3 | Qwen 3 VL 32B Instruct | 47.3 | #606 |
| 4 | Qwen 3 VL 8B Instruct | 42.4 | #1029 |
| 5 | LLaVA-OV-72B | 41.7 | #1879 |
| 6 | Qwen 2.5 VL 72B Instruct | 40.6 | #841 |
| 7 | InternVL3.5-38B | 38.6 | #1021 |
| 8 | Qwen 2.5 VL 32B Instruct | 36.7 | #1062 |
| 9 | Qwen 3 VL 4B Instruct | 35.6 | #1332 |
| 10 | Qwen 2.5 VL 7B Instruct | 35.1 | #1902 |
| 11 | InternVL3.5-14B | 34.1 | #1504 |
| 12 | LLaVA-OV-7B | 33.5 | #4475 |
| 13 | InternVL3.5-4B | 31.6 | #1930 |
| 14 | LLaVA-OV-0.5B | 30.5 | #4352 |
| 15 | Qwen 3 VL 2B Instruct | 30.3 | #2367 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=syncr · How It Works · Data refreshed daily, snapshot 2026-11-21.