EvoGUI-Bench - Temporal Ordering: leaderboard
Metric: Exact ordering accuracy (%; share of 1,000 sets of 3 to 5 shuffled screenshots from one session put back in chronological order; on EvoGUI-Bench, 3,000 zero-shot visual question answering probes mined from Mind2Web and WebLINX web trajectories across 120 domains, screenshots at max_image_side 512, temperature 0; higher is better). Source: arxiv.org. Saturation forecast: Around January 2028. 28 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Flash (Preview) | 58.3 |
| 2 | Doubao-Seed-2.0-Pro-260215 | 44.4 |
| 3 | GPT-5.4 | 42.1 |
| 4 | Kimi K2.6 | 34.4 |
| 5 | Qwen 3.5 397B A17B | 34.4 |
| 6 | MiMo-V2.5 | 34.2 |
| 7 | Kimi K2.5 | 33.4 |
| 8 | Qwen 3.6 35B A3B | 33.3 |
| 9 | Qwen 3.5 122B A10B | 32.8 |
| 10 | GLM-5V Turbo | 31.9 |
| 11 | Qwen 3.5 35B A3B | 31.1 |
| 12 | GLM-4.5V | 30.9 |
| 13 | Qwen 3 VL 235B A22B | 30.9 |
| 14 | GPT-5 Mini | 30.3 |
| 15 | Qwen 3.5 27B | 29.7 |
Interactive version: theaggregate.ai/benchmark?slug=evogui-bench-temporal-ordering · How It Works · Data refreshed daily, snapshot 2026-09-29.