Skip to main content

How to evaluate a langgraph graph

Key concepts

langgraph is a library for building stateful, multi-actor applications with LLMs, used to create agent and multi-agent workflows. Evaluating langgraph graphs can be challenging because a single invocation can involve many LLM calls, and which LLM calls are made may depend on the outputs of preceding calls. In this guide we will focus on the mechanics of how to pass graphs and graph nodes to evaluate() / aevaluate(). For evaluation techniques and best practices when building agents head to the langgraph docs.

End-to-end evaluations

The most common type of evaluation is an end-to-end one, where we want to evaluate the final graph output for each example input.

Define a graph

Lets construct a simple ReACT agent to start:

from typing import Annotated, Literal, TypedDict

from langchain.chat_models import init_chat_model
from langchain_core.tools import tool
from langgraph.graph import END, START, StateGraph
from langgraph.prebuilt import ToolNode
from langgraph.graph.message import add_messages

class State(TypedDict): # Messages have the type "list". The 'add_messages' function # in the annotation defines how this state key should be updated # (in this case, it appends messages to the list, rather than overwriting them)
messages: Annotated[list, add_messages]

# Define the tools for the agent to use
@tool
def search(query: str) -> str:
"""Call to surf the web.""" # This is a placeholder, but don't tell the LLM that...
if "sf" in query.lower() or "san francisco" in query.lower():
return "It's 60 degrees and foggy."
return "It's 90 degrees and sunny."

tools = [search]
tool_node = ToolNode(tools)
model = init_chat_model("claude-3-5-sonnet-latest").bind_tools(tools)

# Define the function that determines whether to continue or not
def should_continue(state: State) -> Literal["tools", END]:
messages = state['messages']
last_message = messages[-1] # If the LLM makes a tool call, then we route to the "tools" node
if last_message.tool_calls:
return "tools" # Otherwise, we stop (reply to the user)
return END

# Define the function that calls the model

def call_model(state: State):
messages = state['messages']
response = model.invoke(messages) # We return a list, because this will get added to the existing list
return {"messages": [response]}

# Define a new graph
workflow = StateGraph(State)

# Define the two nodes we will cycle between
workflow.add_node("agent", call_model)
workflow.add_node("tools", tool_node)

# Set the entrypoint as 'agent'
# This means that this node is the first one called
workflow.add_edge(START, "agent")

# We now add a conditional edge
workflow.add_conditional_edges( # First, we define the start node. We use 'agent'. # This means these are the edges taken after the 'agent' node is called.
"agent", # Next, we pass in the function that will determine which node is called next.
should_continue,
)

# We now add a normal edge from 'tools' to 'agent'.
# This means that after 'tools' is called, 'agent' node is called next.
workflow.add_edge("tools", 'agent')

# Finally, we compile it!
# This compiles it into a LangChain Runnable,
# meaning you can use it as you would any other runnable.
# Note that we're (optionally) passing the memory when compiling the graph
app = workflow.compile()

Create a dataset

Let's create a simple dataset of questions and expected responses:

from langsmith import Client

questions = [
"what's the weather in sf",
"whats the weather in san fran",
"whats the weather in tangier"
]
answers = [
"It's 60 degrees and foggy.",
"It's 60 degrees and foggy.",
"It's 90 degrees and sunny.",
]

ls_client = Client()

dataset = ls_client.create_dataset(
"weather agent",
inputs=[{"question": q} for q in questions],
outputs=[{"answers": a} for a in answers],
)

Create an evaluator

And a simple evaluator:

Requires langsmith>=0.2.0

judge_llm = init_chat_model("gpt-4o")

async def correct(outputs: dict, reference_outputs: dict) -> bool:
instructions = (
"Given an actual answer and an expected answer, determine whether"
" the actual answer contains all of the information in the"
" expected answer. Respond with 'CORRECT' if the actual answer"
" does contain all of the expected information and 'INCORRECT'"
" otherwise. Do not include anything else in your response."
)
# Our graph outputs a State dictionary, which in this case means
# we'll have a 'messages' key and the final message should
# be our actual answer.
actual_answer = outputs["messages"][-1].content
expected_answer = reference_outputs["answer"]
user_msg = (
f"ACTUAL ANSWER: {actual_answer}"
f"\n\nEXPECTED ANSWER: {expected_answer}"
)
response = await judge_llm.ainvoke(
[
{"role": "system", "content": instructions},
{"role": "user", "content": user_msg}
]
)
return response.content.upper() == "CORRECT"

Run evaluations

Now we can run our evaluations and explore the results. We'll just need to wrap our graph function so that it can take inputs in the format they're stored on our example:

Evaluating with async nodes

If all of your graph nodes are defined as sync functions then you can use evaluate or aevaluate. If any of you nodes are defined as async, you'll need to use aevaluate

Requires langsmith>=0.2.0

from langsmith import aevaluate

def example_to_state(inputs: dict) -> dict:
return {"messages": [{"role": "user", "content": inputs['question']}]}

# We use LCEL declarative syntax here.
# Remember that langgraph graphs are also langchain runnables.
target = example_to_state | app

experiment_results = await aevaluate(
target,
data="weather agent",
evaluators=[correct],
max_concurrency=4, # optional
experiment_prefix="claude-3.5-baseline", # optional
)

Evaluating intermediate steps

Often it is valuable to evaluate not only the final output of an agent but also the intermediate steps it has taken. What's nice about langgraph is that the output of a graph is a state object that often already carries information about the intermediate steps taken. Usually we can evaluate whatever we're interested in just by looking at the messages in our state. For example, we can look at the messages to assert that the model invoked the 'search' tool upon as a first step.

Requires langsmith>=0.2.0

def right_tool(outputs: dict) -> bool:
tool_calls = outputs["messages"][1].tool_calls
return bool(tool_calls and tool_calls[0]["name"] == "search")

experiment_results = await aevaluate(
target,
data="weather agent",
evaluators=[correct, right_tool],
max_concurrency=4, # optional
experiment_prefix="claude-3.5-baseline", # optional
)

If we need access to information about intermediate steps that isn't in state, we can look at the Run object. This contains the full traces for all node inputs and outputs:

Custom evaluators

See more about what arguments you can pass to custom evaluators in this how-to guide.

from langsmith.schemas import Run, Example

def right_tool_from_run(run: Run, example: Example) -> dict:
# Get documents and answer
first_model_run = next(run for run in root_run.child_runs if run.name == "agent")
tool_calls = first_model_run.outputs["messages"][-1].tool_calls
right_tool = bool(tool_calls and tool_calls[0]["name"] == "search")
return {"key": "right_tool", "value": right_tool}

experiment_results = await aevaluate(
target,
data="weather agent",
evaluators=[correct, right_tool_from_run],
max_concurrency=4, # optional
experiment_prefix="claude-3.5-baseline", # optional
)

Running and evaluating individual nodes

Sometimes you want to evaluate a single node directly to save time and costs. langgraph makes it easy to do this. In this case we can even continue using the evaluators we've been using.

node_target = example_to_state | app.nodes["agent"]

node_experiment_results = await aevaluate(
node_target,
data="weather agent",
evaluators=[right_tool_from_run],
max_concurrency=4, # optional
experiment_prefix="claude-3.5-model-node", # optional
)

Reference code

Click to see a consolidated code snippet
from typing import Annotated, Literal, TypedDict

from langchain.chat_models import init_chat_model
from langchain_core.tools import tool
from langgraph.graph import END, START, StateGraph
from langgraph.prebuilt import ToolNode
from langgraph.graph.message import add_messages
from langsmith import Client, aevaluate

# Define a graph

class State(TypedDict): # Messages have the type "list". The 'add_messages' function # in the annotation defines how this state key should be updated # (in this case, it appends messages to the list, rather than overwriting them)
messages: Annotated[list, add_messages]

# Define the tools for the agent to use

@tool
def search(query: str) -> str:
"""Call to surf the web.""" # This is a placeholder, but don't tell the LLM that...
if "sf" in query.lower() or "san francisco" in query.lower():
return "It's 60 degrees and foggy."
return "It's 90 degrees and sunny."

tools = [search]
tool_node = ToolNode(tools)
model = init_chat_model("claude-3-5-sonnet-latest").bind_tools(tools)

# Define the function that determines whether to continue or not

def should_continue(state: State) -> Literal["tools", END]:
messages = state['messages']
last_message = messages[-1] # If the LLM makes a tool call, then we route to the "tools" node
if last_message.tool_calls:
return "tools" # Otherwise, we stop (reply to the user)
return END

# Define the function that calls the model

def call_model(state: State):
messages = state['messages']
response = model.invoke(messages) # We return a list, because this will get added to the existing list
return {"messages": [response]}

# Define a new graph
workflow = StateGraph(State)

# Define the two nodes we will cycle between
workflow.add_node("agent", call_model)
workflow.add_node("tools", tool_node)

# Set the entrypoint as 'agent'
# This means that this node is the first one called
workflow.add_edge(START, "agent")

# We now add a conditional edge
workflow.add_conditional_edges( # First, we define the start node. We use 'agent'. # This means these are the edges taken after the 'agent' node is called.
"agent", # Next, we pass in the function that will determine which node is called next.
should_continue,
)

# We now add a normal edge from 'tools' to 'agent'.
# This means that after 'tools' is called, 'agent' node is called next.
workflow.add_edge("tools", 'agent')

# Finally, we compile it!
# This compiles it into a LangChain Runnable,
# meaning you can use it as you would any other runnable.
# Note that we're (optionally) passing the memory when compiling the graph
app = workflow.compile()

questions = [
"what's the weather in sf",
"whats the weather in san fran",
"whats the weather in tangier"
]
answers = [
"It's 60 degrees and foggy.",
"It's 60 degrees and foggy.",
"It's 90 degrees and sunny.",
]

# Create a dataset
ls_client = Client()

dataset = ls_client.create_dataset(
"weather agent",
inputs=[{"question": q} for q in questions],
outputs=[{"answers": a} for a in answers],
)

# Define evaluators

async def correct(outputs: dict, reference_outputs: dict) -> bool:
instructions = (
"Given an actual answer and an expected answer, determine whether"
" the actual answer contains all of the information in the"
" expected answer. Respond with 'CORRECT' if the actual answer"
" does contain all of the expected information and 'INCORRECT'"
" otherwise. Do not include anything else in your response."
)
# Our graph outputs a State dictionary, which in this case means
# we'll have a 'messages' key and the final message should
# be our actual answer.
actual_answer = outputs["messages"][-1].content
expected_answer = reference_outputs["answer"]
user_msg = (
f"ACTUAL ANSWER: {actual_answer}"
f"\n\nEXPECTED ANSWER: {expected_answer}"
)
response = await judge_llm.ainvoke(
[
{"role": "system", "content": instructions},
{"role": "user", "content": user_msg}
]
)
return response.content.upper() == "CORRECT"


def right_tool(outputs: dict) -> bool:
tool_calls = outputs["messages"][1].tool_calls
return bool(tool_calls and tool_calls[0]["name"] == "search")

# Run evaluation

experiment_results = await aevaluate(
target,
data="weather agent",
evaluators=[correct, right_tool],
max_concurrency=4, # optional
experiment_prefix="claude-3.5-baseline", # optional
)

Was this page helpful?


You can leave detailed feedback on GitHub.