OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Agent Frameworks

AG2

AG2 is an open-source Python framework for building protocol-driven AI-agent applications and coordinating multiple agents around LLM-based tasks. It provides agent, tool-use, model-provider, human-in-the-loop and network-orchestration concepts, with the current top-level package imported as `ag2`.

Independent OpenViST project profile, based on linked sources.

AG2 — OpenViST project profile in Agent Frameworks.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use AG2 when an LLM application needs programmable agents, cooperation between multiple agents, tool use or human-in-the-loop control rather than a single model call.

Who it is for

  • Python developers building LLM applications
  • Teams designing multi-agent workflows
  • Researchers developing or studying agentic AI systems
  • Developers integrating agents with tools or MCP-related systems

Documented uses

  • Build protocol-driven AI-agent applications
  • Coordinate multiple agents through network-based orchestration
  • Connect agents with tools and large language models
  • Include human participation in agent workflows
  • Maintain or migrate AutoGen-style applications through AG2 Classic
Before you choose

Requirements & limits

AG2 is available as the `ag2` package on PyPI. The repository documents Python 3.10 or later and provider-specific installation extras. No account, pricing or service-availability details are established here.

  • Choose the current ag2 package for the protocol-driven framework and import ag2 API.
  • Applications using import autogen, ConversableAgent or GroupChat belong to AG2 Classic, which has separate documentation and installation.
  • Migration from AG2 Classic requires modification because the agent model, orchestration and imports changed.
  • MCP-related use is likely compatible, but the available material does not document a specific integration or setup.
  • Provider selection affects which installation extra and standard environment variable are needed.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

COMPARE OPTIONS

Compare options for Agent orchestration

These projects document a common task and share a project role. This is a source-based comparison, not an execution benchmark or a guarantee of interchangeability.

ProjectDocumented purposePlatformAccess and requirements
AG2Build protocol-driven AI-agent applicationsCLI, Linux, macOS, Self-hostedAG2 is available as the `ag2` package on PyPI. The repository documents Python 3.10 or later and provider-specific installation extras. No account, pricing or service-availability details are established here.
AgeniteDefine typed tools with JSON Schema validation and error handling.CLI
VeldraCreate an agent with a policy, tools, skills, knowledge base and reasoning method.Self-hostedThe documented deployment is self-hosted and uses a sign-in gate after a first-run administrator account is created.
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
3 October 2026
Original website title
ag2ai/ag2 · Official homepage title · retained in its source language
How we review sourcesSuggest a correction