Skip to main content

​
OpenAI Integration

The TypeScript SDK provides automatic instrumentation for all OpenAI API calls using OpenTelemetry.

​
How It Works

When you initialize with instrumentModules: { OpenAI }:
  1. The SDK instruments the OpenAI client using OpenTelemetry
  2. Every API call is automatically captured as an Event
  3. Token usage and costs are calculated automatically
  4. Multimodal inputs (text + images) are handled seamlessly

​
Setup

import * as lai from 'lucidicai';
import OpenAI from 'openai';

// Initialize with manual instrumentation (ESM-friendly)
await lai.init({
  sessionName: 'OpenAI Demo',
  instrumentModules: { OpenAI }
});

const openai = new OpenAI({ apiKey: process.env.OPENAI_API_KEY });

// All calls are now automatically tracked!

​
What Gets Captured

For every OpenAI API call:
  • Input: Messages, prompts, and system instructions
  • Model: The model used (gpt-4, gpt-3.5-turbo, etc.)
  • Output: Complete response including function calls
  • Token Usage: Input and output token counts
  • Cost: Calculated based on current pricing
  • Timing: Request duration
  • Images: Automatically uploaded when using vision models

​
Examples

​
Basic Chat Completion

const response = await openai.chat.completions.create({
  model: 'gpt-4',
  messages: [
    { role: 'system', content: 'You are a helpful assistant.' },
    { role: 'user', content: 'Explain quantum computing in simple terms.' }
  ]
});

// Automatically tracked with:
// - Model: gpt-4
// - Input/output tokens
// - Cost calculation
// - Response content

​
Streaming Responses

const stream = await openai.chat.completions.create({
  model: 'gpt-4',
  messages: [{ role: 'user', content: 'Write a haiku about coding' }],
  stream: true
});

for await (const chunk of stream) {
  process.stdout.write(chunk.choices[0]?.delta?.content || '');
}

// Streaming is tracked with accumulated tokens

​
Vision Models

const response = await openai.chat.completions.create({
  model: 'gpt-4-vision-preview',
  messages: [{
    role: 'user',
    content: [
      { type: 'text', text: 'What is in this image?' },
      { 
        type: 'image_url', 
        image_url: { 
          url: 'data:image/jpeg;base64,/9j/4AAQSkZJRg...' 
        } 
      }
    ]
  }]
});

// Images are automatically:
// - Extracted from the request
// - Uploaded to Lucidic
// - Linked to the event

​
Function Calling

const response = await openai.chat.completions.create({
  model: 'gpt-4',
  messages: [{ role: 'user', content: 'What is the weather in SF?' }],
  tools: [{
    type: 'function',
    function: {
      name: 'get_weather',
      description: 'Get weather for a location',
      parameters: {
        type: 'object',
        properties: {
          location: { type: 'string' }
        }
      }
    }
  }]
});

// Tool calls are captured in the event

​
Embeddings

const embedding = await openai.embeddings.create({
  model: 'text-embedding-3-small',
  input: 'The quick brown fox jumps over the lazy dog'
});

// Embedding calls are tracked with:
// - Model used
// - Input text
// - Token count
// - Cost

​
Minimal Example

import * as lai from 'lucidicai';
import OpenAI from 'openai';

// Just these 3 lines for full tracking!
await lai.init({ instrumentModules: { OpenAI } });
const openai = new OpenAI();
const response = await openai.chat.completions.create({
  model: 'gpt-4',
  messages: [{ role: 'user', content: 'Hello!' }]
});

// That's it! Check your dashboard for the tracked event

​
Cost Tracking

The SDK includes built-in pricing for all OpenAI models:
ModelInput CostOutput Cost
gpt-4$0.03/1K$0.06/1K
gpt-4-turbo$0.01/1K$0.03/1K
gpt-3.5-turbo$0.0005/1K$0.0015/1K
o1-preview$0.015/1K$0.06/1K
o1-mini$0.003/1K$0.012/1K
Costs are automatically calculated and attached to each event.

​
Advanced Configuration

​
Custom OpenAI Client Options

await lai.init({ providers: ['openai'] });

const OpenAI = (await import('openai')).default;
const openai = new OpenAI({
  apiKey: process.env.OPENAI_API_KEY,
  organization: 'org-...',
  baseURL: 'https://api.openai.com/v1',
  timeout: 60000,
  maxRetries: 3
});

// All options work with tracking

​
Error Handling

try {
  const response = await openai.chat.completions.create({
    model: 'gpt-4',
    messages: [{ role: 'user', content: 'Hello!' }]
  });
} catch (error) {
  // Errors are also tracked as events
  console.error('OpenAI error:', error);
}

​
Troubleshooting

​
Events Not Appearing

  1. Check import order - OpenAI must be imported AFTER lai.init()
  2. Verify initialization - Ensure instrumentModules included the OpenAI module

​
Incorrect Costs

The SDK uses built-in pricing data. For custom models or pricing: 
// Manually create event with custom cost
await lai.createEvent({
  description: 'Custom model call',
  model: 'ft:gpt-3.5-turbo-custom',
  costAdded: 0.05
});

​
See Also