Adoption Radar

Adopt means use it. Trial means try it. Assess means watch it. Hold means avoid for now. Each tool links the proof.

What the rings mean

  • Adopt. Use it. Stable, updated in the last 6 months, lots of people use it.
  • Trial. Try it. Still young (before 1.0), or quiet for 6 to 18 months, or few users yet.
  • Assess. Watch it. Brand new, less than a year old.
  • Hold. Avoid for now. Dead, deprecated, or no update in 18 months.

Dot shape shows the license: ● open source, ■ free or free tier, ◆ paid. Every tool links the release page or notice that puts it in its ring.

The same tools as a table
ToolFamilyCategoryRingLicenseEasy to useBig real systems
LangGraph Stateful graphs for LLM agents AI & agents Agent frameworks Adopt Open source 40 20
LangChain Model, tool and retriever building blocks for LLM apps AI & agents Agent frameworks Adopt Open source 40 0
CrewAI Role-based multi-agent crews and flows in Python AI & agents Agent frameworks Adopt Open source 60 0
AutoGen Microsoft's multi-agent conversation framework AI & agents Agent frameworks Hold Open source 40 10
AG2 Community fork of AutoGen 0.2 for multi-agent apps AI & agents Agent frameworks Adopt Open source 40 10
Microsoft Agent Framework Successor to Semantic Kernel and AutoGen for .NET and Python AI & agents Agent frameworks Adopt Open source 50 20
Semantic Kernel Microsoft SDK for AI plugins in C#, Python and Java AI & agents Agent frameworks Adopt Open source 40 10
Mastra TypeScript framework for agents, workflows and RAG AI & agents Agent frameworks Adopt Open source 90 20
OpenAI Agents SDK Lightweight agents, handoffs and guardrails from OpenAI AI & agents Agent frameworks Trial Open source 40 10
Claude Agent SDK The Claude Code agent loop as a library AI & agents Agent frameworks Trial Open source 50 0
Google ADK Google's Agent Development Kit for Python, Java, Go and TS AI & agents Agent frameworks Adopt Open source 50 20
Pydantic AI Type-safe Python agent framework from the Pydantic team AI & agents Agent frameworks Adopt Open source 60 10
LlamaIndex Data and retrieval framework for LLM agents AI & agents Agent frameworks Trial Open source 30 10
Haystack Pipelines for RAG and agents from deepset AI & agents Agent frameworks Adopt Open source 40 20
DSPy Program and optimize LM pipelines instead of prompting AI & agents Agent frameworks Adopt Open source 40 0
Vercel AI SDK TypeScript toolkit for LLM apps, streaming UI and agents AI & agents Agent frameworks Adopt Open source 60 10
smolagents Hugging Face's minimal code-writing agents AI & agents Agent frameworks Adopt Open source 50 10
Agno Python agents and teams with a runtime and control plane AI & agents Agent frameworks Adopt Open source 50 20
Letta Stateful agents with long-term memory, formerly MemGPT AI & agents Agent frameworks Trial Open source 50 20
Temporal Durable execution for long-running agent workflows AI & agents Agent frameworks Adopt Open source 60 70
Inngest Durable steps, queues and AgentKit for serverless code AI & agents Agent frameworks Adopt Free tier, paid plans 80 20
Restate Durable execution engine for agents and services AI & agents Agent frameworks Adopt Open source 60 50
Model Context Protocol Open protocol and SDKs connecting agents to tools and data AI & agents Agent tools and MCP Adopt Open source 50 0
MCP Inspector Visual test and debug tool for MCP servers AI & agents Agent tools and MCP Adopt Open source 90 0
Docker MCP Toolkit Run catalog MCP servers as containers behind one gateway AI & agents Agent tools and MCP Trial Free tier, paid plans 50 30
Smithery Registry and hosting for MCP servers AI & agents Agent tools and MCP Hold Free tier, paid plans 40 0
MCP Registry Official community registry of public MCP servers AI & agents Agent tools and MCP Trial Open source 20 0
FastMCP Pythonic framework for MCP servers and clients AI & agents Agent tools and MCP Adopt Open source 80 0
Composio Hosted tool integrations and auth for agents AI & agents Agent tools and MCP Trial Free tier, paid plans 40 10
skills.sh Directory and CLI for installing agent skills AI & agents Agent tools and MCP Assess Open source 50 0
Stagehand Browser automation SDK for agents, by Browserbase AI & agents Agent tools and MCP Adopt Open source 50 0
Playwright MCP MCP server that drives a browser through Playwright AI & agents Agent tools and MCP Trial Open source 60 0
E2B Cloud sandboxes for running agent-generated code AI & agents Agent tools and MCP Adopt Free tier, paid plans 50 20
Claude Code Anthropic's agentic coding tool for the terminal and IDEs AI & agents Coding agents Adopt Paid 90 0
Codex OpenAI's coding agent for terminal, IDE and cloud AI & agents Coding agents Trial Open source 80 0
Cursor AI-first code editor forked from VS Code AI & agents Coding agents Adopt Free tier, paid plans 60 0
Gemini CLI Google's open-source terminal agent for Gemini AI & agents Coding agents Trial Open source 80 0
GitHub Copilot GitHub's AI pair programmer, agent mode and cloud agent AI & agents Coding agents Adopt Free tier, paid plans 60 0
Devin Desktop (formerly Windsurf) Cognition's agentic IDE, renamed from Windsurf AI & agents Coding agents Trial Free tier, paid plans 60 0
Cline Open-source coding agent for VS Code and the terminal AI & agents Coding agents Adopt Open source 70 0
Roo Code VS Code agent forked from Cline, now discontinued AI & agents Coding agents Hold Open source 50 0
Aider Git-aware AI pair programming in the terminal AI & agents Coding agents Trial Open source 60 0
Continue Open-source IDE agent and CI checks with any model AI & agents Coding agents Adopt Open source 60 0
Zed Fast Rust editor with built-in agent panel and ACP AI & agents Coding agents Adopt Open source 80 0
Warp Agentic terminal and development environment AI & agents Coding agents Adopt Free tier, paid plans 50 0
Amp Coding agent for terminal and editors, from Sourcegraph AI & agents Coding agents Trial Free tier, paid plans 50 0
OpenCode Open-source terminal coding agent for any model AI & agents Coding agents Adopt Open source 90 0
goose Local, extensible open-source AI agent built on MCP AI & agents Coding agents Adopt Open source 60 0
Kiro AWS's spec-driven agentic IDE and CLI AI & agents Coding agents Trial Free tier, paid plans 50 0
Jules Google's asynchronous cloud coding agent AI & agents Coding agents Trial Free tier, paid plans 30 10
Devin Cognition's autonomous cloud software engineer AI & agents Coding agents Adopt Paid 40 20
Paradev (formerly Superset) Desktop app to run many coding agents in parallel worktrees AI & agents Coding agents Trial Free tier, paid plans 40 0
Conductor Mac app to run parallel Claude Code and Codex agents AI & agents Coding agents Trial Free, closed source 40 0
Ollama Run open models locally behind a simple API AI & agents Local models Trial Open source 70 10
LM Studio Desktop app to download and serve local models AI & agents Local models Trial Free, closed source 70 0
llama.cpp C/C++ inference for GGUF models on CPU and GPU AI & agents Local models Trial Open source 50 10
vLLM High-throughput LLM serving engine AI & agents Local models Trial Open source 30 40
LocalAI Self-hosted OpenAI-compatible API for many model types AI & agents Local models Adopt Open source 60 20
Docker Model Runner Pull and run models as OCI artifacts with Docker AI & agents Local models Adopt Open source 60 30
Jan Open-source desktop ChatGPT alternative that runs offline AI & agents Local models Trial Open source 70 0
MLX LM Run and fine-tune LLMs on Apple silicon with MLX AI & agents Local models Trial Open source 40 0
Langfuse Open-source LLM tracing, evals and prompt management AI & agents LLM observability Adopt Open source 60 60
LangSmith LangChain's tracing, evals and deployment platform AI & agents LLM observability Adopt Free tier, paid plans 50 20
Arize Phoenix Open-source LLM tracing and evaluation on OpenTelemetry AI & agents LLM observability Adopt Open source 80 30
Helicone Proxy-based LLM logging and AI gateway AI & agents LLM observability Hold Open source 30 10
OpenLLMetry OpenTelemetry instrumentation for LLM libraries AI & agents LLM observability Trial Open source 50 10
Braintrust Evals, logging and prompt playground for AI products AI & agents LLM observability Adopt Free tier, paid plans 40 20
promptfoo CLI for LLM evals and red teaming in CI AI & agents LLM observability Trial Open source 70 0
Opik Open-source LLM tracing and evals from Comet AI & agents LLM observability Adopt Open source 60 50