FAQ
Common questions.
The questions we get most often. If yours isn't here, drop us a line or open a GitHub discussion.
Everything you might ask
An open-source AI agent framework. Drop-in compatible with LangGraph, runs on a Rust engine. Build agents in Python, TypeScript, Go, Java, or Rust — same API in every language.
No. It's a from-scratch Rust implementation with the same API surface. Think of it as LangGraph recompiled as native code, plus circuit breakers, distributed execution, and multi-language SDKs.
Yes. from_langchain_tools() wraps any LangChain BaseTool. as_langchain_runnable() wraps Agentmatic agents as LangChain Runnables — so you keep LangSmith tracing, LCEL composition, and LangServe deployment if you want them.
10–15× faster graph traversal, 70–80× faster channel throughput, 50–80% less memory in our benchmark suite. Production users report 8–12× p95 latency drops on multi-agent workloads. Full numbers at /benchmarks.
MIT license. No platform fees, no per-node pricing, no usage limits. Everything is self-hosted. Optional paid support tiers exist for teams that want SLAs — see /pricing.
OpenAI, Anthropic Claude, Google Gemini, AWS Bedrock, and Ollama out of the box. Any OpenAI-compatible endpoint works (Groq, Together, vLLM, local models).
Different paradigm. CrewAI and AutoGen are role-based delegation. Agentmatic is graph-based state machines — deterministic, parallel, checkpointable. See /compare for head-to-heads.
Yes. npm install @agentmatic/core. Same API as the Python SDK, via napi-rs bindings to the Rust core. Works in Node 20+, Deno, and Bun.
No. The Rust engine is an implementation detail. Most users will only ever write Python or TypeScript code. The Rust crate is there if you want it.
Yes, with Ollama or any local OpenAI-compatible endpoint (vLLM, llama.cpp server). No network calls are made by the framework itself.
Any way you deploy any Python / Node / Go / Java service. Lambda, Cloud Run, k8s, bare metal. The Rust core is statically linked into the SDK binary — no extra runtime to install.
LangGraph Platform's persistence, retry, and distributed execution features are all in our open-source core. The deployment + observability dashboard from Platform is replaced by Agentmatic Studio (also open source) + your existing OpenTelemetry stack.
Yes. agentmatic-studio is a separate package that mounts on top of any running Agentmatic agent. Step debugger, checkpoint timeline, state diff viewer, graph visualization. Open source.
GitHub issues at github.com/neul-labs/agentmatic. Security bugs go through GitHub Security Advisories.
Paid support contracts for teams that want SLAs. We will never relicense or split the codebase. If we disappear, the code is yours under MIT.
Ship your next agent in minutes, not weeks.
MIT licensed. Drop-in for LangGraph. Native SDKs in 5 languages. Battle-tested resilience primitives in the box.