LiteLLM gives every model provider the same call shape, and its proxy adds keys, budgets, rate limits, fallbacks and logging across a whole team. It is the piece you add when several services all need model access.
Reach for it when
- You call more than one provider
- You need per team budgets, keys and rate limits
- You want automatic fallback when a provider fails
Look elsewhere when
- You use exactly one model from one provider
- You already have a gateway
What the lessons will cover
- 01Install and first call
- 02Provider strings
- 03The proxy server
- 04Keys and budgets
- 05Fallbacks and retries
- 06Caching
- 07Logging and observability
- 08Running it in production
Others in infrastructure
E2BSandboxes so agents can run the code they write.
AgentOpsTracing, cost and replay for agents in production.
MastraThe TypeScript agent framework, batteries included.Common questions
Is LiteLLM free to use?
LiteLLM is open source and free to run yourself. You still pay whichever model provider you point it at, and this tutorial is free with no signup.
Do I need to know Python to use LiteLLM?
Basic Python is enough. Functions, dictionaries and imports cover most of what LiteLLM asks of you.
When should I not use LiteLLM?
You use exactly one model from one provider. You already have a gateway.
Lessons for LiteLLM are being written. Meanwhile the LangGraph tutorial covers the same ground: state, tools, loops, memory and human approval. Most of it carries straight over.
Start the LangGraph tutorial →