List your company

GUI Task Automation on WebArena Lite v2

45.5Success RateHiViG
10.34819.47428.637.726Jun 9, 2026

Evaluation results

MethodLinks
HiViGPolicy Backbone=Gemini-3-Flash2026.0645.5
Base agentPolicy Backbone=Gemini-3-Flash2026.0630.5
SE-WSMPolicy Backbone=Gemini-3-Flash2026.0630.5
OpenCUAPolicy Backbone=Gemini-3-Flash2026.0629.9
CGI (32B)Policy Backbone=Gemini-3-Flash2026.0629.9
CGI (8B)Policy Backbone=Gemini-3-Flash2026.0626.6
HiViGPolicy Backbone=Qwen3-VL-32B-Thinking2026.0625.3
GUI-Critic-R1Policy Backbone=Gemini-3-Flash2026.0622.1
CGI (32B)Policy Backbone=Qwen3-VL-32B-Thinking2026.0616.9
OpenCUAPolicy Backbone=Qwen3-VL-32B-Thinking2026.0614.9
GUI-Critic-R1Policy Backbone=Qwen3-VL-32B-Thinking2026.0613.6
Base agentPolicy Backbone=Qwen3-VL-32B-Thinking2026.0613
CGI (8B)Policy Backbone=Qwen3-VL-32B-Thinking2026.0612.3
SE-WSMPolicy Backbone=Qwen3-VL-32B-Thinking2026.0611.7