Open-AutoGLM logo

AI AGENTS

Open-AutoGLM

by zai-org

agentlead-sourcegithub-agent-framework-repositoriessource:github.comphone-agentopen-autoglmautoglm-phone-9bimage-text-to-texttransformersvllmsglangdockergithubhugging-facemodelscope
Visit WebsiteGitHub

Overview

Open phone agent model and framework for using AI to understand smartphone screens and execute automated phone operations.

Details

Open-AutoGLM is described by its GitHub repository as “An Open Phone Agent Model & Framework” for “Unlocking the AI Phone for Everyone.” Related AutoGLM-Phone-9B model pages on Hugging Face and ModelScope describe it as an image-text-to-text model. ModelScope states that Phone Agent uses AutoGLM to understand smartphone screens and execute automated operations, while Hugging Face lists usage snippets for Transformers, vLLM, SGLang, and Docker.

When to Use

Build or evaluate phone-agent workflows that need to understand smartphone screen content and perform automated operations. Experiment with the AutoGLM-Phone-9B image-text-to-text model through Hugging Face, ModelScope, Transformers, vLLM, SGLang, or Docker examples. Research open phone-agent model and framework implementations from the zai-org/Open-AutoGLM repository.

Getting Started

  1. Open the GitHub repository at https://github.com/zai-org/Open-AutoGLM to review the project and framework materials.
  2. Review the Hugging Face model page for zai-org/AutoGLM-Phone-9B and its Transformers, vLLM, SGLang, and Docker usage snippets.
  3. Review the ModelScope page for ZhipuAI/AutoGLM-Phone-9B, including the model description and listed license.
  4. Check the linked arXiv paper 2411.00820 from the Hugging Face model page for research background.

Key Features

  • •Open phone agent model and framework, according to the GitHub repository description.
  • •AutoGLM-Phone-9B image-text-to-text model availability on Hugging Face and ModelScope.
  • •Designed for phone-agent use cases where AutoGLM understands smartphone screens and executes automated operations, according to ModelScope.
  • •Usage examples are listed on Hugging Face for Transformers, vLLM, SGLang, and Docker.

Capabilities

  • •phone-agent-framework
  • •smartphone-screen-understanding
  • •automated-phone-operations
  • •image-text-to-text-model
  • •transformers-compatible
  • •vllm-compatible
  • •sglang-compatible
  • •docker-usage

Last updated May 31, 2026