Multiple Input Graph 🔢

Objectives ✅

Define a more complex AgentState
Create a processing node that performs operations on list data
Set up a LangGraph that processes and outputs computed results.
Invoke the graph with structured inputs and retrieve outputs

from typing import List, TypedDict
 
from langgraph.graph import StateGraph
 
class AgentState(TypedDict):
    values: List[int]
    name: str
    result: str

def process_values(state: AgentState) -> AgentState:
    """This function processes a list of values and returns their sum as a string."""
 
    total = sum(state['values'])
    state['result'] = f"The sum of values for {state['name']} is {total}."
    return state

graph = StateGraph(AgentState)
 
graph.add_node("processor", process_values)
 
# START
graph.set_entry_point("processor")
 
# END
graph.set_finish_point("processor")
 
app = graph.compile()

result = app.invoke({"values": [1, 2, 3, 4, 5], "name": "Alice"})
 
result['result']

"multi input graph"

OUTPUT:
 
'The sum of values for Alice is 15.'

2. Hello World Graph 4. Sequential Graph