List your company

Web Navigation on WebArena Protocol A 812 tasks

66.9Success Rate (SR)GBT-SE
17.39630.24843.155.952Jan 30, 2026

Evaluation results

MethodLinks
GBT-SEbase_system=OpenHands CodeAct2026.0166.9780.2052205
GBT-Basicbase_system=OpenHands CodeAct2026.015376.50.41060230
OpenHands CodeAct (gpt-4o, native)mode=native2026.0119.7—3.410094360
OpenHands CodeAct + Global guardrail onlyconfiguration=Global guardrail only2026.0119.3—0.92088335