EvoGUI-Bench: leaderboard

Metric: EvoGain index (mean over the three probes of the gain over the random baseline normalized by the headroom above it, times 100: temporal ordering exact-match accuracy, inverse action and value joint accuracy and one-step successor accuracy; 100 is perfect, 0 is random and below-random performance is negative, down to about -42; 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 2029. 28 models tracked.

Top models

#ModelScore
1Gemini 3 Flash (Preview)60.4
2Doubao-Seed-2.0-Pro-26021549.8
3Kimi K2.643.9
4Qwen 3.5 122B A10B43.8
5GPT-5.443.4
6MiMo-V2.542.4
7Qwen 3.6 35B A3B42
8GLM-5V Turbo41.9
9Kimi K2.541.6
10Qwen 3.5 35B A3B40.5
11Qwen 3.5 27B39.6
12Qwen 3.5 397B A17B39.5
13Qwen 3.6 27B38.5
14Qwen 3 VL 235B A22B37.5
15GLM-4.5V37.1

Interactive version: theaggregate.ai/benchmark?slug=evogui-bench · How It Works · Data refreshed daily, snapshot 2026-09-29.