LangGraph Checkpoint Aerospike

Store LangGraph checkpoints in Aerospike using the provided AerospikeSaver.

Installation

pip install -U langgraph-checkpoint-aerospike

Usage

1. Bring up Aerospike locally using prebuilt Aerospike Docker Image:

docker run -d --name aerospike -p 3000-3002:3000-3002 container.aerospike.com/aerospike/aerospike-server

2. Point the saver at your cluster (Default):

   • AEROSPIKE_HOST=127.0.0.1
   • AEROSPIKE_PORT=3000
   • AEROSPIKE_NAMESPACE=test

3. Build a LangGraph graph and compile it with the saver as the checkpointer. Passing the same thread_id on a later run resumes from the persisted state:

   from typing import Annotated, TypedDict
   
   import aerospike
   from langgraph.graph import START, END, StateGraph
   from langgraph.graph.message import add_messages
   from langchain_core.messages import HumanMessage
   
   from langgraph.checkpoint.aerospike import AerospikeSaver
   
   # 1. Connect to Aerospike and build the checkpointer.
   client = aerospike.client({"hosts": [("127.0.0.1", 3000)]}).connect()
   checkpointer = AerospikeSaver(client=client, namespace="test")
   
   # 2. Define a minimal LangGraph graph.
   class State(TypedDict):
       messages: Annotated[list, add_messages]
   
   def chatbot(state: State) -> State:
       last = state["messages"][-1].content
       return {"messages": [("assistant", f"You said: {last}")]}
   
   builder = StateGraph(State)
   builder.add_node("chatbot", chatbot)
   builder.add_edge(START, "chatbot")
   builder.add_edge("chatbot", END)
   
   # 3. Compile with the Aerospike checkpointer.
   graph = builder.compile(checkpointer=checkpointer)
   
   # 4. Each thread_id is a separate, persisted conversation.
   config = {"configurable": {"thread_id": "demo"}}
   graph.invoke({"messages": [HumanMessage("hello")]}, config)
   
   # 5. A later call with the same thread_id resumes from the saved state,
   #    even in a brand-new process — the history lives in Aerospike.
   result = graph.invoke({"messages": [HumanMessage("are you still there?")]}, config)
   for message in result["messages"]:
       message.pretty_print()
   

You can also inspect persisted checkpoints directly through the saver, e.g. checkpointer.get_tuple(config) for the latest checkpoint or checkpointer.list(config) to walk a thread's history.