AutoGen / AG2

https://github.com/ag2ai/ag2

Originally Microsoft Research's multi-agent framework — fork branded AG2 (2024-10) led by community. Memory is one optional component (AutoGen Memory cross-listed).

Active · Verified 2026-10-31 · fresh as of 2026-10-31 · 45 cited facts, 9 with a third-party source · 1 recorded change · Suggest a correction

At a glance

Tier
T1
Created
2023 repo fix
Latest release
v1.2.1 (2026-10-21) registry fix
License
Apache-2.0 repo fix
Pricing
free (OSS) repo fix
Funding
searched not found fix
Freshness
2026-10-31 fix
Layer
agent-infra repo fix
Licensing
open-source repo fix
Deployment locus
local repo fix
Platform vs product
platform repo fix
Maturity
ga registry fix
SDK languages
python registry fix

When to use

Optimised for: searched not found

Anti-fit: searched not found

Pros & cons

Pros

searched not found

Cons

searched not found

Claims & capabilities

Originally Microsoft Research AutoGen; forked to AG2 in 2024-10 by community after Microsoft AutoGen renamed; MIT.

Technical surface

API surface
searched not found
Backend storage
not applicable — not a memory product
Deployment
Self-hosted only — library
Embedding model
not applicable — not a memory product
Multi-tenancy
searched not found
MCP
yes — official MCP client support (autogen.mcp.create_toolkit)
A2A
yes — native A2A server + client since AG2 v0.10 (2025-10)
OpenTelemetry
searched not found

Similar systems

Other agent frameworks (no first-party memory layer) in the catalog, ranked by inbound references.

  • Model Context Protocol (MCP spec) T1

    Anthropic-introduced open protocol (Nov 2024) for connecting LLMs to data sources and tools. Donated by Anthropic to the Linux Foundation's Agentic AI Foundation (announced 2025-12-09); now governed as Model Context Protocol, a Series of LF Projects, LLC. Supported across Claude, ChatGPT, VS Code, Cursor and many other clients.

  • LangChain (framework) T1

    Dominant Python/JS framework for LLM apps — tool calling, agents, retrievers, runnables. Memory is one component (LangMem cross-listed); the framework itself is the product.

  • Vercel AI SDK T1

    Vercel's TypeScript LLM SDK — streaming, tool calls, structured outputs, agents. Dominant framework for Next.js LLM apps.

  • LlamaIndex (framework) T1

    Data framework for LLM apps — connectors, indices, query engines, agents, workflows. Memory is one of many subsystems (LlamaIndex Memory cross-listed).

  • OpenAI Agents SDK T1

    OpenAI's official Python framework for agents — handoffs, guardrails, tracing built in. Successor to OpenAI Swarm (experimental); officially supported.

  • LangGraph (orchestration) T1

    LangChain's stateful graph runtime for agents — explicit state, persistence, time-travel debugging, human-in-the-loop. Memory is one persistence target (LangGraph Persistence cross-listed).

Sources (44 cited facts, 9 third-party)

Each fact links to the page it came from; the badge says who published it (how to read sources).

What changed

Changes to this record's lifecycle, licence, release, pricing and hosting facts, from the catalog's git history. Subscribe (Atom)

  1. 2026-10-26
    • Latest release first recorded as “v1.2.1 (2026-10-21)” source
    • Pricing first recorded as “free (OSS)” source
    • Deployment first recorded as “Self-hosted only — library” source
    • MCP support first recorded as “yes — official MCP client support (autogen.mcp.create_toolkit)” source
    • A2A support first recorded as “yes — native A2A server + client since AG2 v0.10 (2025-10)” source
    #258

Related systems

References (14)

  • Agent2Agent Protocol (A2A) integrates with — AG2 docs have an A2A section for serving and calling agents over the A2A protocol. (verified 2026-10-29)
  • Anthropic Claude (foundation models) integrates with — AG2 model configuration docs list Anthropic as a supported model provider with a config example. (verified 2026-10-29)
  • Arize AI integrates with — AG2 telemetry docs point to Arize AX as a managed path for tracing and monitoring AG2 agents. (verified 2026-10-29)
  • Arize Phoenix integrates with — AG2 telemetry docs name Arize Phoenix as the open-source path for inspecting AG2 OpenTelemetry traces. (verified 2026-10-29)
  • Daytona integrates with — AG2 code execution docs list Daytona as a hosted code execution environment (DaytonaEnvironment). (verified 2026-10-29)
  • Google Gemini 3 family integrates with — AG2 model configuration docs list Gemini as a supported model provider with a config example. (verified 2026-10-29)
  • GraphRAG (Microsoft) integrates with — Microsoft AutoGen autogen-ext ships a GraphRAG search tool (autogen_ext.tools.graphrag). (verified 2026-10-29)
  • LangChain (framework) integrates with — Microsoft AutoGen autogen-ext ships a LangChain tool adapter (autogen_ext.tools.langchain). (verified 2026-10-29)
  • Microsoft Semantic Kernel integrates with — Microsoft AutoGen autogen-ext ships a Semantic Kernel model client adapter and tool adapter. (verified 2026-10-29)
  • Mistral Large 2 / Mixtral family integrates with — AG2 model configuration docs list Mistral as a supported model provider with a config example. (verified 2026-10-29)
  • Model Context Protocol (MCP spec) integrates with — AG2 docs describe connecting agents to MCP servers as tools. (verified 2026-10-29)
  • OpenAI GPT family (GPT-5 / GPT-4o / o3 / o4) integrates with — AG2 model configuration docs list OpenAI as a supported model provider with a config example. (verified 2026-10-29)
  • vLLM integrates with — AG2 model configuration docs show configuring a self-hosted vLLM OpenAI-compatible endpoint. (verified 2026-10-29)
  • xAI Grok 4 integrates with — AG2 model configuration docs list xAI Grok as a supported model provider with a config example. (verified 2026-10-29)

Referenced by (21)

  • AgentOps integrates with — AgentOps docs have a AutoGen / AG2 integration page. (verified 2026-10-29)
  • Arcade.dev integrates with — Arcade docs have a AutoGen / AG2 integration page. (verified 2026-10-29)
  • Arize Phoenix integrates with — Arize Phoenix docs have a AutoGen / AG2 integration page. (verified 2026-10-29)
  • Braintrust integrates with — Braintrust docs have a AutoGen / AG2 integration page. (verified 2026-10-29)
  • Comet ML integrates with — Comet's Opik docs have a AutoGen / AG2 integration page. (verified 2026-10-29)
  • Composio integrates with — Composio's SDKs and frameworks page lists a AutoGen / AG2 provider. (verified 2026-10-29)
  • Daytona integrates with — Daytona docs have a guide for running AutoGen / AG2 with Daytona sandboxes. (verified 2026-10-29)
  • HoneyHive integrates with — HoneyHive integrations docs have a Microsoft AutoGen tracing/integration page. (verified 2026-10-29)
  • IBM ContextForge MCP Gateway integrates with — IBM ContextForge docs have a AutoGen integration page. (verified 2026-10-29)
  • Invariant Guardrails integrates with — Invariant Gateway README has a Microsoft AutoGen integration section. (verified 2026-10-29)
  • Langfuse integrates with — Langfuse's integrations directory has a AutoGen / AG2 integration page. (verified 2026-10-29)
  • LangSmith integrates with — LangSmith docs have a tracing integration page for AutoGen / AG2. (verified 2026-10-29)
  • Mem0 integrates with — Mem0 docs have an AutoGen integration page (Build conversational AI agents with AutoGen and Mem0 for context-aware, personalized interactions). (verified 2026-10-29)
  • MongoDB Atlas Vector Search integrates with — MongoDB's Atlas AI Integrations page lists AutoGen / AG2 as an integration (some entries are marked community-maintained). (verified 2026-10-29)
  • OpenInference integrates with — The OpenInference repo README lists an OpenInference instrumentation package for AutoGen / AG2. (verified 2026-10-29)
  • Portkey AI integrates with — Portkey docs have an Autogen agent framework integration page. (verified 2026-10-29)
  • Pydantic Logfire integrates with — Pydantic Logfire integrations docs have a AutoGen page. (verified 2026-10-29)
  • Qdrant integrates with — Qdrant's documentation has a AutoGen / AG2 integration page. (verified 2026-10-29)
  • Together AI integrates with — Together AI docs third-party/agent integrations: AutoGen (AG2). (verified 2026-10-29)
  • Weights & Biases integrates with — W&B Weave docs have a AutoGen / AG2 integration page. (verified 2026-10-29)
  • Zep & Graphiti integrates with — Zep docs have an AG2 integration page (zep-ag2 package). (verified 2026-10-29)