AI agent framework / multi-agent orchestration
google/adk-js is a code-first TypeScript toolkit for building, evaluating, and deploying AI agents and multi-agent workflows on Node.js and the web.
Same problem, same approach. Swapping one for another is a config change, not a rewrite.
TypeScript peer with its own agent console; choose it if you prefer VoltAgent's tooling and observability over ADK's Google integrations.
AI Agent Engineering Platform built on an Open Source TypeScript AI Agent Framework
TypeScript agent framework; choose it if you want a TS-native alternative with its own dev tooling and integrations.
Mastra is the modern TypeScript framework for AI-powered applications and agents.
Solves the same problem with a different architecture or at a different layer. Expect to rewrite the integration.
Python multi-agent conversation framework; choose it if your team is Python-first and you want Microsoft's research-oriented agent model.
A programming framework for agentic AI
Python/.NET framework; choose it when you need .NET-native agent orchestration or Microsoft's newer agent abstractions.
A framework for building, orchestrating and deploying AI agents and multi-agent workflows with support for Python and .NET.
Python multi-agent harness; choose it if you want an opinionated production runtime instead of a TypeScript SDK.
Multi-Agent Harness for Production AI
Python batteries-included wrapper; choose it for rapid prototyping with minimal boilerplate, at the cost of TypeScript type safety.
PraisonAI 🦞 — Hire a 24/7 AI Workforce. Stop writing boilerplate and start shipping autonomous self-improving agents that research, plan, code, and execute tasks. Deployed in 5 lines of code with built-in memory, RAG, and support for 100+ LLMs.
TypeScript agent-network platform; choose it when agents must collaborate across a network rather than inside one app.
OpenAgents - AI Agent Networks for Open Collaboration
Java/Spring framework; choose it if your backend is Java and you want Spring-native agent abstractions.
Agentic AI Framework for Java Developers
Python research framework built on CAMEL; choose it for task-automation experiments rather than production integration.
🦉 OWL: Optimized Workforce Learning for General Multi-Agent Assistance in Real-World Task Automation
Python multi-agent software-company framework; choose it for research on agent collaboration rather than product integration.
🌟 The Multi-Agent Framework: First AI Software Company, Towards Natural Language Programming
Python framework for building and running agent platforms; choose it if you want a Python-first alternative with built-in runtime management.
Build, run, and manage agent platforms.
Python graph-based orchestration library; choose it when you need explicit stateful graph control over multi-agent workflows.
Build resilient agents.
Python lightweight multi-agent framework; choose it if you want OpenAI's minimal SDK and are not tied to TypeScript.
A lightweight, powerful framework for multi-agent workflows
C#/.NET SDK; choose it when integrating LLM agents into .NET applications rather than Node.js.
Integrate cutting-edge LLM technology quickly and easily into your apps
Python agent harness built on LangGraph; choose it if you want a batteries-included agent with planning and sub-agents.
The batteries-included agent harness.
Python stateful-agent platform; choose it when long-term memory and self-improvement are the priority over TypeScript tooling.
Platform for stateful agents: AI with advanced memory that can learn and self-improve over time.
Does the same job, but ships as an app. Useful to a person, not swappable into a codebase.
Java enterprise agent platform with visual workflow tools; choose it when you want a managed low-code platform instead of writing code.
Enterprise-grade, commercial-friendly agentic workflow platform for building next-generation SuperAgents.
A local-first agent OS/hub application; choose it if you want a runnable orchestrator rather than a library to embed.
Agent OS: keep specialist agents in a hub, spin up a temporary orchestrator per task. Local-first, works with any model.
Visual low-code LLM app platform; choose it when non-developers need to build and deploy agents through a web workspace.
Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or self-hosted, so teams move from prototype to production without rebuilding the stack.
A ready-to-run super-assistant app; choose it if you want an out-of-the-box personal agent rather than a framework.
Open-source super AI assistant & Agent Harness. Plans tasks, runs tools and skills, self-evolves with memory and knowledge. Multi-agent, multi-model, multi-channel. Lightweight, extensible, one-line install. (formerly chatgpt-on-wechat)
TypeScript visual agent-development platform; choose it when you need a drag-and-drop builder rather than a code-first SDK.
An AI agent development platform with all-in-one visual tools, simplifying agent creation, debugging, and deployment like never before. Coze your way to AI Agent creation.
Python dockerized personal agent app; choose it if you want a ready-to-run autonomous assistant, not an embeddable library.
Agent Zero AI framework
Python LLMOps platform with RAG and workflow UI; choose it when you need an integrated enterprise platform rather than an agent SDK.
BISHENG is an open LLM devops platform for next generation Enterprise AI applications. Powerful and comprehensive features include: GenAI workflow, RAG, Agent, Unified model management, Evaluation, SFT, Dataset Management, Enterprise-level System Management, Observability and more.
These projects were analysed and named adk-js among their alternatives. The relationship is not symmetric — how adk-js rates them is a separate judgement, made when adk-js is analysed in its own right.
calls adk-js “Choose for a TypeScript-first agent framework from Google if your stack is JavaScript/TypeScript.”
🦉 OWL: Optimized Workforce Learning for General Multi-Agent Assistance in Real-World Task Automation
calls adk-js “When you want a TypeScript code-first toolkit for building and deploying agents with fine-grained control.”
BISHENG is an open LLM devops platform for next generation Enterprise AI applications. Powerful and comprehensive features include: GenAI workflow, RAG, Agent, Unified model management, Evaluation, SFT, Dataset Management, Enterprise-level System Management, Observability and more.
calls adk-js “Google's code-first TypeScript agent toolkit; choose if you are in JavaScript/TypeScript and want official Google ADK support.”
Agentic AI Framework for Java Developers
calls adk-js “when you prefer a TypeScript code-first agent toolkit from Google.”
An Autonomous LLM Agent for Complex Task Solving
calls adk-js “Choose for a TypeScript code-first agent toolkit from Google, trading Python ecosystem for TS type safety and deployment flexibility.”
agentUniverse is a LLM multi-agent framework that allows developers to easily build multi-agent applications.