VisualWebArena — leaderboard

VisualWebArena: Measures browser, desktop, mobile, or GUI agents operating in interactive environments.

Metric: Success Rate (%). Source: jykoh.com. Status: saturation imminent. 32 models tracked.

Top models

#ModelScore
1Gemini 2.5 Flash54
2GPT-552.9
3GLM-4 32B48.7
4GPT-4o19.78
5Gemini 1.5 Pro11.98
6GPT-47.25
7Gemini 1.5 Flash6.59
8Gemini 1.0 Pro6.04
9GPT-3.52.2
10Mixtral 8x7B1.76
11Llama 2 70B1.1

Interactive version: theaggregate.ai/benchmark?slug=visualwebarena · How the rankings work · Data refreshed daily, snapshot 2026-07-22.