ChronoVision - Event Photo Ordering: leaderboard
Event photo ordering on ChronoVision (2026), a temporal reasoning benchmark for vision-language models: the percent of 1,200 pairs of same-category event photos in original color for which the model says correctly which one was taken earlier. Higher is better.
Metric: Accuracy (%) on the 1,200 SPEED event-photo pairs in original color: say which of two same-category photos was taken earlier; zero-shot; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 10 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Pro | 86.25 | #266 |
| 2 | Qwen 3 VL 235B A22B Instruct | 77.83 | #522 |
| 3 | GPT-5.2 | 77 | #176 |
| 4 | Qwen 3 VL 8B Instruct | 76.42 | #981 |
| 5 | Qwen 3 VL 4B Instruct | 68.92 | #1279 |
| 6 | GLM-4.1V-9B (Thinking) | 67.67 | #1060 (GLM-4.1V-9B) |
| 7 | Qwen 2.5 VL 7B Instruct | 66.08 | #1833 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=chronovision-event-photo-ordering · How It Works · Data refreshed daily, snapshot 2026-11-13.