AG2

Most enterprises don’t have an AI strategy problem, they have an AI execution problem. Even with proven ROI from successful pilots, agents often remain fragmented and unable to integrate with core business processes. That changes today with AG2.ai. Powered by AgentOS, we transform fragmented experiments into a coordinated, production-ready AI workforce. The winners of 2026 won't be the companies with the most AI pilots, they'll be the ones with the most scalable orchestration. Ready to build your AI workforce? 👉 ag2.ai #EnterpriseAI #AgenticAI #OpenSource

AG2 v0.11.3 & v0.11.4 are here! 🚀 🎉 AG2 Beta — ground-up redesign for production agents: streaming, multi-provider LLM support, typed tools with DI, middleware pipeline, first-class testing 🐟 TinyFish agentic web scraper 🔍 QuickResearchTool for parallel web research 🛡️ Security hardening: shell injection, path traversal, credential redaction 🔄 Mistral V2 + Gemini streaming 🙌 Thanks to our contributors! Release: github.com/ag2ai/ag2/rele… #AG2 #AI #MultiAgentSystems #AIAgents #OpenSource #Beta

Most teams don't lack AI. They lack working AI workflows. McKinsey's 2025 State of AI: 80% of orgs use AI somewhere, but only ~20% redesigned workflows where value is created. Ideas aren't the bottleneck. Durable execution is. Studio inside AgentOS closes that gap. Describe your workflow naturally. CaptainAgent builds the first agent graph. Test it live, iterate visually on canvas (triggers, agents, conditions, human review), then export AG2 code. One workflow that "should already be automated"? Build + test it now: app.ag2.ai #ag2 #agentos #workflows #captainagent

Introducing AG2 Beta - a new development track inside AG2. A ground-up redesign of the agent experience with a production-focused feature set. Simplified API, streaming-first, async-first, event-driven, with dependency injection, and middleware. Developed in parallel, so the current AG2 framework remains fully maintained and supported. Beta agents already work inside existing AG2 group chats and workflows. 5 LLM providers. Powerful new multi-agent orchestrations, observability, MCP, A2A, and AG-UI coming next. We're just getting started, find out more here: docs.ag2.ai/latest/docs/bl… #ag2 #oss #buildinpublic #opensource

Company due diligence is still largely manual. Analysts jump between websites, founder profiles, investor pages, and press coverage to piece together a company story. AG2 enables structured multi-agent research workflows instead of single prompts. Agents can also invoke tools when needed — like TinyFish for crawling sites and extracting data. Full breakdown: ag2.ai/blog/due-dilig…

We taught the playground to build agentic workflows for you. CaptainAgent turns conversation into AG2 orchestration. Describe what you want, it designs the agents, picks the pattern, and generates the code. Modify it conversationally. Copy or download the code and you're done. Try it now: playground.ag2.ai #AI #AG2 #AgenticAI #OpenSource #Agents #CaptainAgent

Latest drop in our open source AG2 examples repo: 📬 Email agent — archive, trash, list emails from chat 🔬 GPT Researcher — 7-stage pipeline, live browser UI, streams in real time 🖥️ AG-UI demos — AG2 streaming to the browser, no React needed (3 examples) ☁️ GCP Starter Pack — AG2 to Cloud Run in one command, A2A protocol out of the box Beginner-friendly to production-ready. ⭐ github.com/ag2ai/build-wi…

47% of enterprises based major decisions on AI hallucinations. (Deloitte, 2025) We've teamed up with @assafelovic to fix this. GPT Researcher now runs inside AG2 — eight specialized agents for deep web research, with human review before writing begins. github.com/ag2ai/build-wi…

AG2 at @daytonaio compute conference! Drop by for a chat or some💫shiny swag if you are around!

We built a playground for multi-agent AI. Watch agents orchestrate, hand off, use tools, and collaborate — live in your browser. Pick a pattern. Hit Run. See what multi-agent AI actually looks like. Also, tell us what else you want to play with! playground.ag2.ai #AI #AG2 #AgenticAI #OpenSource #Agents #Playground

AG2 v0.11.2 – Squashing bugs, sharpening tools. 🔧 🔧 Gemini parallel tool calls + structured output fixed ⚙️ Custom HTTP headers for OpenAI & Azure OpenAI 🪲 Async tool execution & agent handoff fixes 📚 New CopilotKit quickstart + AG-UI docs refresh 🙌 Thanks to our contributors! Release: github.com/ag2ai/ag2/rele… #AG2 #AI #MultiAgentSystems #AIAgents #OpenSource

@ataiiam Let's go agent builders !🚀🚀🚀

AG2 🤝 AG-UI Protocol AG2 has been fully integrated with the AG-UI Protocol! For a long time, there was (only) a Chat UI integration with AG-UI. But I'm happy to announce that anyone building an AG2 agent can now bring it into their frontend applications and build beautiful, fullstack agentic experiences. This includes: - Chat UI - Generative UI - State synchronization - Human-in-the-loop - Frontend Tool Calls ...and more! All the building blocks you need to take your agents from the backend and into a fullstack agentic application. A big thanks to the @ag2oss team for building out this full integration! Resources below 👇🏼

79% of companies use AI agents. Only 20% trust them. Close the trust gap a UI for users to work closely with them. We've teamed up with @CopilotKit, adding native AG-UI support — an open protocol for agent-to-user interaction. Streaming UI, tool visualization, human-in-the-loop — standardized, not custom-built. Get started: ag2.ai/blog/ag-ui

@Tiny_Fish Thrilled to be a partner here supporting all builders! Let's go! 🚀 🚀

$2M seed pool. 9 weeks. 15+ partners. 100% virtual. Introducing the TinyFish Accelerator to fund builders creating autonomous agents for the Agentic Web. Backed by Mango Capital. Apply Now: tinyfish.ai/accelerator What you walk into: → Build in public from Day 1 → Free credits from partner companies like MongoDB, ElevenLabs, Fireworks ai, v0 by Vercel, Google for Startups, and more → Direct engineering support → Weekly business mentorship → Demo Day directly with the investor → High-fidelity tech community The earlier you apply, the more runway you get. #TinyFishAccelerator #BuildInPublic

Hundreds of AI agent demos are about to flood your timeline. Not pitch decks. Not mockups. Working agents. We joined @Tiny_fish's $2M accelerator as a partner because we want our infra in the hands of the sharpest agent builders, before anyone else. What we're giving the cohort: → Generously Free Credits → Direct line to our engineers Some of these teams will be our biggest customers in a year. Applications are rolling: tinyfish.ai/accelerator #TinyFishAccelerator #BuildInPublic

🚀 AG2 v0.11.1 is live! 🖥️ AG-UI Message Streaming – Real-time agent responses in your AG-UI frontends with new event-based streaming architecture. 🌊 A2A Streaming – Full streaming support for Agent-to-Agent communication! LLM text streaming now flows through A2A for real-time responses from remote agents, both server and client-side. 📡 OpenAI – Get more out of OpenAI's latest Responses API with stateful conversations, built-in tools (web search, image gen, apply_patch), reasoning model support, and enhanced cost tracking. Docs: 🛠️🪲 Bug fixes and maintenance #AG2 #AI #MultiAgentSystems #AIAgents #OpenSource #AGUI

🚀 AG2 v0.10.5 is live! 🤖 GPT 5.2 Codex Models Support – Added support for OpenAI's GPT 5.2 Codex models, bringing enhanced coding capabilities to your agents. 🐚 GPT 5.1 Shell Tool Support – The Responses API now supports the shell tool, enabling agents to interact with command-line interfaces for filesystem diagnostics, build/test flows, and complex agentic coding workflows. Check out the blogpost: Shell Tool and Multi-Inbuilt Tool Execution. 🔬 RemyxCodeExecutor – New code executor for research paper execution, expanding AG2's capabilities for scientific and research workflows. Check out the updated code execution documentation: Code Execution. Documentation 📔 New guide: Google Vertex AI deployments 📝 New blogpost: Gemini Thinking Config 📝 New blogpost: Shell Tool and Multi-Inbuilt Tool Execution 🔧 Fixed orchestrations basic concepts documentation. 🛠️🔒 Security and bug fixes Explore the new capabilities here 👇 docs.ag2.ai/latest/ github.com/ag2ai/ag2/rele… #AG2 #AI #MultiAgentSystems #AIAgents #OpenSource

Exactly! We are witnessing the transition of software-as-a-tool to software-as-a-teammate. At AG2, we share this vision for the future of software. Apps are evolving into skills or tools in an Agent’s library. The value shifts entirely to the Orchestration Layer, the intelligence that selects, coordinates, and executes the right tools to deliver results. As apps fade into the background, Agentic Workflows are emerging as a new Operating System. We’re building toward a world where the prompt becomes the primary interface for work, automation, and decision-making.

🚀 AG2 v0.10.4 is live! 🕹️ Step-through Execution - A powerful new orchestration feature (run_iter and run_group_chat_iter) that allows developers to pause and step through agent workflows event-by-event. This enables granular debugging, human-in-the-loop validation, and precise control over the execution loop. ☁️ AWS Bedrock "Thinking" & Reliability - significant upgrades to the Bedrock client: Reliability: Added built-in support for exponential backoff and retries, resolving throttling issues on the Bedrock Converse API. Advanced Config: Added support for additionalModelRequestFields, enabling advanced model features like Claude 3.7 Sonnet's "Thinking Mode" and other provider-specific parameters directly via BedrockConfigEntry. 💰 Accurate Group Chat Cost Tracking - A critical enhancement to cost observability. Previously, group chats might only track the manager or the last agent; this update ensures costs are now correctly aggregated from all participating agents in a group chat session. 🤗 HuggingFace Model Provider - Added a dedicated guide and support documentation for integrating the HuggingFace Model Provider, making it easier to leverage open-source models. 🐍 Python 3.14 Readiness - Added devcontainer.json support for Python 3.14, preparing the development environment for the next generation of Python. 📚 Documentation & Blogs - Comprehensive new resources including: Logging Events: A deep dive into tracking and debugging agent events. MultiMCPSessionManager: Guide on managing multiple Model Context Protocol sessions. Apply Patch Tool: Tutorial on using the patch application tools. 📔 Documentation updates 🛠️🪲 Security and bug fixes Explore the new capabilities here 👇 docs.ag2.ai/latest/ github.com/ag2ai/ag2/rele… #AG2 #AI #MultiModal #A2AProtocol #MultiAgentSystems #AIAgents #OpenSource