omniparser

Last updated: May 1, 2026. Microsoft OmniParser V2 is a vision-based screen parser that turns a UI screenshot into structured, LLM-readable elements (bounding boxes plus icon captions). Pair it with a vision LLM and you have the perception layer for a "computer-use" agent. This guide is a clean,

Last updated April 2026 — refreshed for OmniParser v2.0.1, the CVE-2025-55322 patch, and the current OmniTool stack. Microsoft OmniParser is the screen-parsing layer that turns ordinary multimodal LLMs into GUI agents: feed it a screenshot and it returns a JSON list of every interactable element with bounding boxes, function

Microsoft's OmniParser V2 is an advanced AI model designed to interpret screen elements from screenshots, predicting the coordinates and descriptions of all elements. When combined with Large Language Models (LLMs), it enables AI to interact with any application through vision, similar to human interaction. Why V2 Over V1?