Aegra: Self-Hosted
AI Agent Backend
LangGraph power without vendor lock-in. Run AI agents on your infrastructure.

Why Self-Hosted
Pay only for your infrastructure, not per-node
Use the same LangGraph SDK with full control
Keep your AI conversations and models on your infrastructure
Aegra gives you complete control over your AI agent infrastructure. Self-host with confidence, scale without limits, and integrate seamlessly with your existing LangGraph workflows.
Based on LangChain pricing as of February 2026. An enterprise tier with self-hosting is also available at custom pricing.
| Feature | LangSmith Deployments | Aegra |
|---|---|---|
| Deploy agents | Local dev only (Free), paid cloud (Plus) | Free, unlimited |
| Custom auth | Not available (Free), available (Plus) | Python handlers (JWT/OAuth/Firebase) |
| Self-hosted | Enterprise only (license key required) | Always (Apache 2.0) |
| Own database | Managed only (Free/Plus), BYO (Enterprise) | Bring your own Postgres |
| Tracing | LangSmith only | Any OTLP backend (Langfuse, Phoenix, etc.) |
| Data residency | LangChain cloud (Free/Plus), yours (Enterprise) | Your infrastructure |
| SDK | LangGraph SDK | Same LangGraph SDK |
Enterprise-Grade Features,
Developer-First Design
Production-ready features for self-hosted AI agent infrastructure that scales with your needs.
Agent Protocol Compliant
Implements the open-source Agent Protocol specification for standardized LLM agent APIs and seamless integration.
LangGraph SDK Compatible
Drop-in replacement for LangGraph Platform. Use your existing client code without any modifications.
PostgreSQL Persistence
Durable conversation storage with checkpoints, streaming responses, and network resilience built-in.
Docker Ready
5-minute deployment with Docker Compose. Includes database migrations, health checks, and monitoring endpoints.
Custom Authentication
Extensible auth framework supporting JWT, OAuth, Firebase, or no-auth for development environments.
Multi-LLM Support
Works with OpenAI, Anthropic, Together AI, and other providers. Switch models without changing your agent code.
What others say
"You save my life. I am doing a state of art chatbot for mental health and the pay for execution node killed my project. Aegra lets me run everything on my own servers."
Rodrigo T.
Healthcare AI Developer
"I'll gradually migrate my existing LangGraph framework to use Aegra. Finally found something that actually works as advertised."
Daniel C.
ML Engineering Lead
"If it worked with Agent Chat UI, I would switch to aegra right away. Turns out it does work, so we made the switch last month."
BeepSoft Team
Enterprise Development
"I love the idea of this repo, very willing to be your test subject. Been running it in production for 3 weeks now without issues."
Elijah S.
Startup Founder
Proudly Open Source
Built by developers, for developers who value transparency and control.