Langfuse

Integrate Agno with Langfuse to send traces and gain insights into your agent's performance.

Integrating Agno with Langfuse

Langfuse provides a robust platform for tracing and monitoring AI model calls. By integrating Agno with Langfuse, you can utilize OpenInference and OpenLIT to send traces and gain insights into your agent's performance.

Set OpenAI Key

Set your OPENAI_API_KEY as an environment variable. You can get one from OpenAI.

export OPENAI_API_KEY=sk-***

Prerequisites

  1. Install Dependencies

    Ensure you have the necessary packages installed:

    uv pip install agno openai langfuse opentelemetry-sdk opentelemetry-exporter-otlp openinference-instrumentation-agno openlit yfinance

  2. Setup Langfuse Account

    • Either self-host or sign up for an account at Langfuse.
    • Obtain your public and secret API keys from the Langfuse dashboard.

  3. Set Environment Variables

    Configure your environment with the Langfuse API keys:

    export LANGFUSE_PUBLIC_KEY="your-public-key"
export LANGFUSE_SECRET_KEY="your-secret-key"

Run each alternative in a fresh Python process. Each configures the global tracer provider. The x-langfuse-ingestion-version=4 header enables Langfuse's current ingestion path; direct OTLP ingestion without it can be delayed. See the Langfuse OpenTelemetry guide.

Sending Traces to Langfuse

Example: Using Langfuse with OpenInference

Instrument your Agno agent with OpenInference and send traces to Langfuse.

import base64
import os

from agno.agent import Agent
from agno.models.openai import OpenAIResponses
from agno.tools.yfinance import YFinanceTools
from openinference.instrumentation.agno import AgnoInstrumentor
from opentelemetry import trace as trace_api
from opentelemetry.exporter.otlp.proto.http.trace_exporter import OTLPSpanExporter
from opentelemetry.sdk.trace import TracerProvider
from opentelemetry.sdk.trace.export import SimpleSpanProcessor

# Set environment variables for Langfuse
LANGFUSE_AUTH = base64.b64encode(
    f"{os.getenv('LANGFUSE_PUBLIC_KEY')}:{os.getenv('LANGFUSE_SECRET_KEY')}".encode()
).decode()
os.environ["OTEL_EXPORTER_OTLP_ENDPOINT"] = "https://us.cloud.langfuse.com/api/public/otel"
os.environ["OTEL_EXPORTER_OTLP_HEADERS"] = (
    f"Authorization=Basic {LANGFUSE_AUTH},x-langfuse-ingestion-version=4"
)

# Configure the tracer provider
tracer_provider = TracerProvider()
tracer_provider.add_span_processor(SimpleSpanProcessor(OTLPSpanExporter()))
trace_api.set_tracer_provider(tracer_provider=tracer_provider)

# Start instrumenting agno
AgnoInstrumentor().instrument()

# Create and configure the agent
agent = Agent(
    name="Stock Price Agent",
    model=OpenAIResponses(id="gpt-5.2"),
    tools=[YFinanceTools()],
    instructions="You are a stock price agent. Answer questions in the style of a stock analyst.",
    debug_mode=True,
)

# Use the agent
agent.print_response("What is the current price of Tesla?")

Example: Using Langfuse with OpenLIT

Use Langfuse via OpenLIT to trace model calls.

import base64
import os

from agno.agent import Agent
from agno.models.openai import OpenAIResponses
from agno.tools.hackernews import HackerNewsTools
from opentelemetry.exporter.otlp.proto.http.trace_exporter import OTLPSpanExporter
from opentelemetry.sdk.trace import TracerProvider
from opentelemetry.sdk.trace.export import SimpleSpanProcessor
from opentelemetry import trace

# Set environment variables for Langfuse
LANGFUSE_AUTH = base64.b64encode(
    f"{os.getenv('LANGFUSE_PUBLIC_KEY')}:{os.getenv('LANGFUSE_SECRET_KEY')}".encode()
).decode()
os.environ["OTEL_EXPORTER_OTLP_ENDPOINT"] = "https://us.cloud.langfuse.com/api/public/otel"
os.environ["OTEL_EXPORTER_OTLP_HEADERS"] = (
    f"Authorization=Basic {LANGFUSE_AUTH},x-langfuse-ingestion-version=4"
)

# Configure the tracer provider
trace_provider = TracerProvider()
trace_provider.add_span_processor(SimpleSpanProcessor(OTLPSpanExporter()))
trace.set_tracer_provider(trace_provider)

# Initialize OpenLIT instrumentation
import openlit
# OpenLIT reuses the global tracer provider configured above
openlit.init(disable_batch=True, disable_metrics=True, disable_events=True)

# Create and configure the agent
agent = Agent(
    model=OpenAIResponses(id="gpt-5.2"),
    tools=[HackerNewsTools()],
    markdown=True,
    debug_mode=True,
)

# Use the agent
agent.print_response("What are the top stories on HackerNews right now?")

Notes

  • Environment Variables: Ensure your environment variables are correctly set for the API keys and OTLP endpoint.
  • Data Regions: Adjust the OTEL_EXPORTER_OTLP_ENDPOINT for your data region or local deployment as needed. Available regions include:
    • https://us.cloud.langfuse.com/api/public/otel for the US region
    • https://cloud.langfuse.com/api/public/otel for the EU region
    • http://localhost:3000/api/public/otel for local deployment