Search Results for "visual-"

...Notably, it was the first known project to implement a multimodal model capable of viewing and controlling a computer screen. The framework supports features like Optical Character Recognition (OCR) and Set-of-Mark (SoM) prompting to enhance visual grounding capabilities. It is designed to be compatible with macOS, Windows, and Linux (with X server installed), and is released under the MIT license.

...The latest version, Agent S3, surpasses human-level performance on the OSWorld benchmark, demonstrating state-of-the-art results in complex multi-step computer tasks. Agent S combines powerful foundation models (such as GPT-5) with grounding models like UI-TARS to translate visual inputs into precise executable actions. It supports flexible deployment via CLI, SDK, or cloud, and integrates with multiple model providers including OpenAI, Anthropic, Gemini, Azure, and Hugging Face endpoints. With optional local code execution, reflection mechanisms, and compositional planning, Agent S provides a scalable and research-driven framework for building advanced computer-use agents.

...It aims to create an “AI phone agent” that can perceive on-screen content, reason about user goals, and execute sequences of taps, swipes, and text input via automated device control interfaces like ADB, enabling hands-off completion of multi-step tasks such as navigating apps, filling forms, and more. Unlike traditional automation scripts that depend on brittle heuristics, Open-AutoGLM uses pretrained large language and vision-language models to interpret visual context and natural language instructions, giving the agent robust adaptability across apps and interfaces.

airda(Air Data Agent) is a multi-smart body for data analysis, capable of understanding data development and data analysis needs, understanding data, generating data-oriented queries, data visualization, machine learning and other tasks of SQL and Python codes.