Sending Traces

Tracing is a fundamental feature in LLM application development, enabling developers to gain deep insights into their apps’ performance and behavior. By capturing detailed information such as input and output tokens, response times, model parameters, and user feedback, tracing allows you to optimize your LLM applications effectively. This guide will walk you through the process of sending traces from your application to Langtrace Cloud, helping you leverage this powerful tool for analysis and improvement.

​

Step 1: Set Up Langtrace Cloud

To begin sending traces, you’ll need to set up your Langtrace Cloud account:

1. Sign up by going to this link.
2. After signing up, create a new Project. Projects are containers for storing traces and metrics generated by your application. If you have only one application, creating 1 project will suffice.
3. Generate an API key. This key will be used to authenticate your application with Langtrace Cloud.
   You may also create new projects and generate API keys for each of them later.

​

Step 2: Install and Initialize the SDK

To start sending traces, you need to install the Langtrace SDK and initialize it with your API key. Follow these steps:

# Install the SDK
pip install langtrace-python-sdk

# Import it into your project

from langtrace_python_sdk import langtrace # Must precede any llm module imports
langtrace.init(api_key = '<LANGTRACE_API_KEY>')

Replace <LANGTRACE_API_KEY> with the API key you generated in Step 1.

​

Step 3: Create and Send Traces

Once the SDK is initialized, Langtrace will automatically capture traces from supported LLM integrations.

​

Example: Sending Traces from an OpenAI Application

Here’s an example of how to send traces from an LLM application using the OpenAI API. This example demonstrates how to use the Langtrace SDK to send traces from an OpenAI completion request.

pip install --upgrade langtrace-python-sdk openai

npm install @langtrase/typescript-sdk openai

from langtrace_python_sdk import langtrace, with_langtrace_root_span

from openai import OpenAI

langtrace.init(
  api_key="<YOUR API KEY>")

@with_langtrace_root_span()
def example():
  client = OpenAI()
  response = client.chat.completions.create(
      model="gpt-3.5-turbo",
      messages=[
          {
              "role": "system",
              "content": "How many states of matter are there?"
          }
      ],
  )
  print(response.choices[0].message.content)

example()

Run the code snippet as shown below:

export OPENAI_API_KEY="<YOUR OPENAI API KEY>"
python main.py

After setting up Langtrace and running your application, your traces will be automatically sent to Langtrace Cloud, where you can view and analyze them on the dashboard.

​

Next Steps

Now that you’ve set up Langtrace and are sending traces, you can:

1. View Your Traces: Log in to your Langtrace Cloud account to view and analyze the traces you’ve sent.

2. Customize Your Tracing: Explore adding custom attributes or spans to capture more specific data about your LLM operations.

3. Integrate with Your Workflow: Consider how you can use the insights from your traces to optimize your LLM applications.

4. Explore Advanced Features: Check out our other guides to learn about features like:

• Adding custom attributes
• Grouping traces
• Integrating user feedback

​

Troubleshooting Common Issues

If you encounter problems while setting up or using Langtrace, here are solutions to some common issues:

If you’re still experiencing issues after trying these solutions, please contact our support team.