Data as of Jul 25, 2026 · Based on 2,718,867 AI responses across 9,511 prompts · See how Parse measures this
CogAgentgithub.com · Low evidenceCogAgent is an open-source vision-language model (VLM) designed as a GUI agent for automated interaction with graphical user interfaces. It supports bilingual Chinese and English input via screenshots and natural language, achieving state-of-the-art results in GUI perception, reasoning, and task execution across multiple benchmarks.
Parse Score