@owlmetric/tracker

A Node.js wrapper for logging LLM traces directly to Owlmetric, bypassing the proxy. Track usage metrics and monitor your AI applications with support for OpenAI, Anthropic, Google Gemini, and more.

Features

• Direct Provider Integration - Wrap your existing OpenAI, Anthropic, or Gemini clients
• Usage Tracking - Automatic token counting and cost tracking
• Streaming Support - Full support for streaming completions
• OpenTelemetry Compatible - Use with existing OTEL infrastructure
• Next.js Ready - Built-in support for Next.js applications
• Type Safe - Full TypeScript support

Installation

npm install @owlmetric/tracker

Quick Start

Direct Provider Integration

OpenAI

import OpenAI from "openai";
import { createTrackedClient } from "@owlmetric/tracker";

const client = createTrackedClient(OpenAI, {
  apiKey: process.env.OPENAI_API_KEY,
  owlmetricToken: process.env.OWLMETRIC_OPENAI_TOKEN!,
});

// Regular completion
const completion = await client.chat.completions.create({
  model: "gpt-4",
  messages: [{ role: "user", content: "Hello!" }],
});

// Streaming completion
const stream = await client.chat.completions.create({
  model: "gpt-4",
  stream: true,
  messages: [{ role: "user", content: "Count from 1 to 5" }],
});

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

Anthropic

import Anthropic from "@anthropic-ai/sdk";
import { createTrackedClient } from "@owlmetric/tracker";

const client = createTrackedClient(Anthropic, {
  apiKey: process.env.ANTHROPIC_API_KEY,
  owlmetricToken: process.env.OWLMETRIC_ANTHROPIC_TOKEN!,
});

// Regular completion
const completion = await client.messages.create({
  model: "claude-sonnet-4-20250514",
  max_tokens: 50,
  messages: [{ role: "user", content: "Hello!" }],
});

// Streaming completion
const streamResponse = await client.messages.create({
  model: "claude-sonnet-4-20250514",
  max_tokens: 100,
  messages: [{ role: "user", content: "Count from 1 to 5" }],
  stream: true,
});

for await (const event of streamResponse) {
  if (event.type === "content_block_delta" && event.delta?.type === "text_delta") {
    process.stdout.write(event.delta.text);
  }
}

Google Gemini

import { GoogleGenAI } from "@google/genai";
import { createTrackedClient } from "@owlmetric/tracker";

const client = createTrackedClient(GoogleGenAI, {
  apiKey: process.env.GEMINI_API_KEY,
  owlmetricToken: process.env.OWLMETRIC_GEMINI_TOKEN!,
});

// Regular completion
const completion = await client.models.generateContent({
  model: "gemini-2.5-flash",
  contents: "Hello!",
});

// Streaming completion
const stream = await client.models.generateContentStream({
  model: "gemini-2.5-flash",
  contents: "Count from 1 to 5",
});

for await (const chunk of stream) {
  const text = typeof chunk.text === "function" ? chunk.text : chunk.text;
  if (text) process.stdout.write(text);
}

Next.js Integration

Installation

npm install @owlmetric/tracker @vercel/otel @opentelemetry/sdk-logs @opentelemetry/api-logs @opentelemetry/instrumentation

1. Setup Instrumentation

Create instrumentation.ts in your Next.js root directory:

import { registerOTel } from "@vercel/otel";
import { OwlMetricTraceExporter } from "@owlmetric/tracker/owlmetric_trace_exporter";

export function register() {
  registerOTel({
    serviceName: "next-app",
    traceExporter: new OwlMetricTraceExporter(),
  });
}

2. API Route with AI SDK

// app/api/chat/route.ts
import { openai } from "@ai-sdk/openai";
import { anthropic } from "@ai-sdk/anthropic";
import { google } from "@ai-sdk/google";
import { streamText } from "ai";

export const maxDuration = 30;

export async function POST(req: Request) {
  const { messages } = await req.json();

  const result = streamText({
    model: openai("gpt-4o"), // or anthropic("claude-3-haiku-20240307"), google("gemini-2.0-flash")
    messages,
    experimental_telemetry: {
      isEnabled: true,
      metadata: {
        xOwlToken: process.env.OWLMETRIC_TOKEN, // Your Owlmetric token
      },
    },
  });

  return result.toDataStreamResponse();
}

OpenTelemetry Integration (NodeJS)

Basic Setup

import { OwlMetricTraceLogger } from "@owlmetric/tracker/owlmetric_trace_logger";
import { createOpenAI } from "@ai-sdk/openai";
import { generateText, streamText } from "ai";

// Initialize the logger
const logger = new OwlMetricTraceLogger();
logger.init();

const ai = createOpenAI({
  apiKey: process.env.OPENAI_API_KEY,
});

const model = ai("gpt-4");

// Use with telemetry
const { textStream } = streamText({
  model: model,
  prompt: "Hello, world!",
  experimental_telemetry: {
    isEnabled: true,
    metadata: {
      userId: "user-123",
      sessionId: "session-456",
      xOwlToken: process.env.OWLMETRIC_TOKEN,
    },
  },
});

for await (const textPart of textStream) {
  console.log(textPart);
}

Environment Variables

Set up your environment variables:

# Provider API Keys
OPENAI_API_KEY=your_openai_key
ANTHROPIC_API_KEY=your_anthropic_key
GEMINI_API_KEY=your_gemini_key

# Owlmetric Tokens (one per provider)
OWLMETRIC_OPENAI_TOKEN=pt_your_openai_token
OWLMETRIC_ANTHROPIC_TOKEN=pt_your_anthropic_token
OWLMETRIC_GEMINI_TOKEN=pt_your_gemini_token

API Reference

createTrackedClient(ClientClass, options)

Creates a tracked version of your AI client.

Parameters:

• ClientClass: The original client class (OpenAI, Anthropic, GoogleGenAI)
• options: Configuration object including API key and Owlmetric token

Returns: Tracked client instance with same API as the original

OwlMetricTraceExporter

OpenTelemetry trace exporter for Owlmetric.

import { OwlMetricTraceExporter } from "@owlmetric/tracker";

const exporter = new OwlMetricTraceExporter();

OwlMetricTraceLogger

Logger for initializing OpenTelemetry tracing.

import { OwlMetricTraceLogger } from "@owlmetric/tracker";

const logger = new OwlMetricTraceLogger();
logger.init();

Supported Providers

•  OpenAI (GPT-3.5, GPT-4, GPT-4o)
•  Anthropic (Claude 3, Claude 3.5 Sonnet)
•  Google Gemini (Gemini Pro, Gemini 2.0 Flash)
•  DeepSeek
•  AI SDK compatible providers

License

SEE LICENSE IN LICENSE.md

Support

For issues and support, please visit our GitHub repository.