LangGraph (Python)

Low-level orchestration for stateful, long-running LLM agents. LangGraph models your agent as a graph of nodes and edges with explicit state, durable checkpoints, and first-class human-in-the-loop.

Install

pip install langgraph langchain-anthropic

Version

v1.2.11 · August 2026 · Python 3.10 – 3.14

Best for

Cycles, conditional routing, multi-agent coordination, HITL workflows.

Where to start

New to LangGraph? Follow the Zero → Hero path below in order (~3–4 hours). Already building? Use Jump to a topic to land on the chapter you need.

Zero → Hero

A chaptered tutorial that takes you from a first graph to a production-ready agent. Each chapter has its own page with Next → pagination.

1 · Setup & core concepts Install, state, nodes, edges, compilation · 15 min

2 · Your first agent Linear chat, conditional routing, looping, streaming · 20 min

3 · Multi-agent systems Supervisor, parallel workers, handoff · 25 min

4 · Tools ToolNode, custom executors, conditional tool use · 20 min

5 · Memory & persistence Checkpointers, Store, cross-thread memory · 25 min

6 · Streaming & debugging All 7 stream modes, typed StreamPart v2, GraphOutput, token streaming, custom events · 30 min

7 · Human-in-the-loop Interrupts, multi-stage approval, interactive debug · 25 min

8 · Middleware pre_model_hook, post_model_hook, RetryPolicy, TimeoutPolicy, error_handler, ToolNode(handle_tool_errors) · 25 min

9 · Advanced patterns ReAct, Tree-of-Thoughts, reflection, structured output, caching, Functional API · 25 min

10 · Production & troubleshooting RetryPolicy, TimeoutPolicy, CachePolicy, Durability, IsLastStep, Topic channels, Docker, CLI · 30 min

Looking for a reference?

The Zero → Hero chapters are task-oriented tutorials. If you already know the shape of LangGraph and want exact signatures, feature tables, and breaking-change history for a single primitive, jump to the Reference section near the bottom of this page — one page per API (StateGraph, Checkpointers, Store, Functional API, Command/Send, streaming modes). Each is verified line-by-line against the installed langgraph==1.2.11 source.

Jump to a topic

Need one specific concept? Land directly on the chapter that owns it.

State graph

Conditional routing

Supervisor pattern

Parallel workers

ToolNode

Checkpointers

Store (long-term memory)

Type-safe v2 API

Interrupts (HITL)

Middleware & hooks

ReAct pattern

Functional API

Command / Send

IsLastStep / RemainingSteps

Runtime / ToolRuntime injection

Stream modes (v1 / v2)

StateGraph API

Checkpointers matrix

Store (long-term memory)

Prebuilt Nodes

Channels

RemoteGraph

Class & API reference (all symbols)

Reference

The API reference pages below are introspected against the installed langgraph==1.2.11 / langgraph-checkpoint==4.2.0 / langgraph-checkpoint-sqlite==3.0.3 / langgraph-checkpoint-postgres==3.0.5 source. If a signature disagrees with an older blog post or docs page, the source wins.

StateGraph API reference Constructor, add_node (all 4 overloads), add_edge, add_conditional_edges, add_sequence, compile, Runtime, RetryPolicy, CachePolicy, TimeoutPolicy, RunControl, ExecutionInfo.

Checkpointers reference Feature matrix for InMemorySaver, Sqlite, AsyncSqlite, Postgres, AsyncPostgres, ShallowPostgres — plus durability modes and connection strings.

Store (long-term memory) reference BaseStore, InMemoryStore, namespaces, IndexConfig (vector search), TTLConfig, InjectedStore for tools.

Functional API reference @entrypoint, @task, entrypoint.final, previous, injecting Runtime, feature matrix vs StateGraph.

Command / Send reference Command(update, goto, resume, graph), Send for fan-out, interrupt + resume, Command.PARENT, Overwrite.

Streaming modes reference All eight stream_mode values (including tools + ToolCallTransformer), v1 vs v2 typing, StreamPart discriminated union, GraphOutput, subgraphs=True.

Prebuilt nodes reference ToolNode (all constructor params, wrap_tool_call, handle_tool_errors), InjectedState, InjectedStore, ToolRuntime, ToolCallRequest, tools_condition, ToolCallTransformer, ToolCallStream.

Channels reference LastValue, BinaryOperatorAggregate, Topic, EphemeralValue, NamedBarrierValue, AnyValue — concurrent-write semantics, Annotated[] shorthands, fan-in patterns.

Managed values reference IsLastStep and RemainingSteps — read-only state fields injected by the Pregel executor for recursion-limit detection and graceful truncation.

Runtime, ToolRuntime & Managed Values reference Runtime context injection, ToolRuntime for tools, ExecutionInfo, IsLastStep, RemainingSteps — accessing execution metadata and injecting immutable context into nodes and tools.

RemoteGraph reference Call a LangGraph Platform deployment as a local graph — same invoke/stream/get_state surface, interrupt/resume, and subgraph node wiring.

Graceful shutdown reference RunControl, request_drain(), GraphDrained — cooperative superstep-boundary draining for SIGTERM, pre-stop hooks, and timeout-with-checkpoint patterns.

TracePolicy reference Per-node LangSmith tracing control — process_inputs, process_outputs, omit_payload; selectively redact, truncate, or silence individual node spans.

Cache backends reference BaseCache abstract interface and InMemoryCache thread-safe implementation — TTL eviction, custom Redis backends, CachePolicy wiring, and serializer protocol.

Single-page comprehensive guide All chapters in one long page — best for search and printing.

Recipes RAG, support routing, research agent, doc pipelines, long-term memory conversations.

Diagrams 20+ Mermaid architecture diagrams.

Production guide Deployment, scaling, monitoring, disaster recovery.

Advanced error recovery

Observability

Performance optimization

Streaming server (FastAPI)

Verified against installed library

Introspected against langgraph==1.2.11, langgraph-checkpoint==4.2.0, langgraph-prebuilt==1.1.0, langgraph-checkpoint-sqlite==3.0.3, langgraph-checkpoint-postgres==3.0.5 in August 2026. The full class/API surface — every symbol below and many more — is consolidated in the Class & API Reference section of the single-page guide.

Known corrections vs older drafts / third-party guides:

• langgraph.llm_hooks and langgraph.cache.cache_node do not exist. The real middleware API is pre_model_hook/post_model_hook kwargs on langgraph.prebuilt.create_react_agent; see Chapter 8.
• langchain.agents.create_agent(middleware=[...]) and AgentMiddleware exist in the separate langchain package (install with pip install langchain), which is not bundled with langgraph. create_react_agent is deprecated per official LangGraph v1 migration docs in favour of langchain.agents.create_agent.
• AgentState / AgentStatePydantic are deprecated in langgraph.prebuilt (no longer re-exported from the top-level __init__; still importable from langgraph.prebuilt.chat_agent_executor). Deprecation notice targets langchain.agents as migration (requires the langchain package).
• SqliteSaver.from_conn_string(...) and PostgresSaver.from_conn_string(...) are context managers (with ... as saver:), not factories that return a saver directly. Some earlier examples omitted the with; see the Checkpointers reference.
• Interrupt.ns, Interrupt.when, Interrupt.resumable were removed in v0.6; only Interrupt.value and Interrupt.id remain.
• checkpoint_during=False on invoke/stream is deprecated — use durability="exit" (or "sync" / "async").
• ShallowPostgresSaver is deprecated as of langgraph-checkpoint-postgres 2.0.20. Use PostgresSaver + durability="exit".

Install, minimal example

from langgraph.graph import StateGraph, START, END
from langgraph.checkpoint.memory import InMemorySaver
from typing_extensions import TypedDict

class State(TypedDict):
    message: str
    response: str

def process(state: State) -> dict:
    return {"response": f"Processed: {state['message']}"}

builder = StateGraph(State)
builder.add_node("process", process)
builder.add_edge(START, "process")
builder.add_edge("process", END)

graph = builder.compile(checkpointer=InMemorySaver())

result = graph.invoke(
    {"message": "Hello"},
    config={"configurable": {"thread_id": "user-1"}},
)
print(result)

Ready for the full walk-through? Start Chapter 1 →

---

Revision history

Full per-volume history now lives in git — this table covers meaningfully distinct versions only, oldest to newest.

Date	Version	Changes
2025-11	1.0.3	Initial guide: node caching, deferred nodes, pre/post model hooks, cross-thread memory, Python 3.13 support.
2026-04-16	1.1.6	Python 3.9 dropped; type-safe v2 streaming/invoke API; Pydantic/dataclass coercion; Python 3.14 support; time-travel fixes.
2026-04-21	1.1.8	Index redesigned into Zero → Hero chaptered tutorial + Jump-to-topic grid.
2026-04-22	1.1.9	Added six source-verified reference pages (StateGraph, Checkpointers, Store, Functional API, Command/Send, Streaming modes); corrected stale guidance (create_react_agent deprecation, from_conn_string context managers, ShallowPostgresSaver deprecation).
2026-05-12	1.2.0	ToolRuntime and ToolCallTransformer added to langgraph.prebuilt.
2026-05-22	1.2.1	Patch release; all core symbols re-verified.
2026-06-15	1.2.5	v3 streaming protocol (beta) documented: GraphRunStream, SubgraphRunStream, native transformers.
2026-06-23	1.2.6	Deprecation of AgentState/AgentStatePydantic/ValidationNode in favor of langchain.agents; LangGraphDeprecatedSinceV11 added.
2026-07-01	1.2.7	TracePolicy, add_node(defer=True), graph visualization/schema-introspection APIs (get_graph(xray=), as_tool(), get_context_jsonschema()) documented.
2026-07-09	1.2.8	Encrypted checkpoint serde (EncryptedSerializer), serde audit hooks documented.
2026-07-20	1.2.9	RunControl/cooperative draining, RemainingSteps, ToolRuntime execution metadata, TimeoutPolicy refinements.
2026-08-01	1.2.10	Patch release; all core symbols re-verified against the installed source.
2026-08-17	1.2.11	Consolidated the 46 “class deep dive” volumes into one source-verified Class & API Reference section on the single-page guide; removed the per-volume pages and changelog entries (full history remains in git); refreshed all version callouts to langgraph==1.2.11.