​
Integrating Agno with Weave by WandB

Weave provides a powerful platform for logging and visualizing model calls. By integrating Agno with Weave, you can track and analyze your agent’s performance and behavior.

​
Prerequisites

  1. Install Dependencies
Ensure you have the necessary packages installed: 
pip install weave
  1. Create a WandB Account
  1. Set Environment Variables
Configure your environment with the WandB API key: 
export WANDB_API_KEY=<your-api-key>

​
Sending Traces to Weave

  • ​
    Example: Using weave.op decorator

This method requires installing the weave package and then utilising @weave.op decorator over any function you wish to automatically trace. This works by creating wrappers around the functions. 
import weave
from agno.agent import Agent
from agno.models.openai import OpenAIChat

# Initialize Weave with your project name
weave.init("agno")

# Create and configure the agent
agent = Agent(model=OpenAIChat(id="gpt-4o"), markdown=True, debug_mode=True)

# Define a function to run the agent, decorated with weave.op()
@weave.op()
def run(content: str):
    return agent.run(content)

# Use the function to log a model call
run("Share a 2 sentence horror story")

  • ​
    Example: Using OpenTelemetry

In this method, we utilize weave’s support for OpenTelemetry based trace logging. This method does not require installing weave Python SDK as a dependency. First, install the required OpenTelemetry dependencies: 
pip install openai openinference-instrumentation-agno opentelemetry-sdk opentelemetry-exporter-otlp-proto-http
This example demonstrates how to instrument your Agno agent with OpenInference and send traces to Weave: 
import base64
import os

from agno.agent import Agent
from agno.models.openai import OpenAIChat
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 the endpoint and headers for Weave
WANDB_BASE_URL = "https://trace.wandb.ai"
PROJECT_ID = "<your-entity>/<your-project>"
OTEL_EXPORTER_OTLP_ENDPOINT = f"{WANDB_BASE_URL}/otel/v1/traces"

# Configure authentication
WANDB_API_KEY = os.getenv("WANDB_API_KEY")
AUTH = base64.b64encode(f"api:{WANDB_API_KEY}".encode()).decode()

headers = {
    "Authorization": f"Basic {AUTH}",
    "project_id": PROJECT_ID,
}

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

# Start instrumenting agno
AgnoInstrumentor().instrument()

# Create and configure the agent
agent = Agent(
    model=OpenAIChat(id="gpt-4o-mini"),
    tools=[YFinanceTools(stock_price=True)],
    instructions="Use tables to display data. Don't include any other text.",
    markdown=True,
    debug_mode=True
)

# Use the agent
agent.print_response("What is the stock price of Apple?", stream=True)

​
Notes

  • Environment Variables: Ensure your environment variables are correctly set for the WandB API key.
  • Project Configuration: Replace <your-entity>/<your-project> with your actual WandB entity and project name for OpenTelemetry setup.
  • Entity Name: You can find your entity name by visiting your WandB dashboard and checking the Teams field in the left sidebar.
  • Method Selection: Use weave.op decorator for simpler setup, or OpenTelemetry for richer logging and better dashboard reporting.
By following these steps, you can effectively integrate Agno with Weave, enabling comprehensive logging and visualization of your AI agents’ model calls.