Skip to Content
Introduction

AiSpendTrack Documentation

Track, analyze, and optimize your LLM API costs in real-time.

AiSpendTrack is a lightweight proxy that gives you complete visibility into your OpenAI and Anthropic API spending. Setup takes 10 minutes.

What You Get

  • 📊 Real-time cost tracking - See every dollar spent as it happens
  • 📈 Trend analysis - 7-day, 30-day, 90-day spending patterns
  • 🎯 Cost breakdown - By model, feature, customer, or endpoint
  • 🚨 Smart alerts - Get notified when costs spike or limits hit
  • 🔒 Privacy-first - Zero prompt logging, metadata only
  • ⚡ Low latency - <300ms overhead on average

Quick Start

Get started in 3 steps:

  1. Sign up →  - Create your free account
  2. Add 2 lines - Update your OpenAI/Anthropic client
  3. Watch costs appear - Real-time dashboard updates
// Before
const openai = new OpenAI({
  apiKey: process.env.OPENAI_API_KEY
});
 
// After (2 lines added)
const openai = new OpenAI({
  apiKey: process.env.OPENAI_API_KEY,
  baseURL: "https://proxy.aispendtrack.com/v1",
  defaultHeaders: {
    "x-aispendtrack-key": "ask_your_key_here"
  }
});

That’s it! Your costs will now appear at app.aispendtrack.com .

What Gets Tracked?

We track metadata only:

  • ✅ Model (gpt-4, claude-3-opus)
  • ✅ Tokens (prompt + completion)
  • ✅ Cost (calculated from pricing)
  • ✅ Latency (response time)
  • ✅ Status (200, 429, 500)

We never track:

  • ❌ Your prompts
  • ❌ AI responses
  • ❌ User data
  • ❌ API keys

Supported Providers

Currently supported:

  • ✅ OpenAI (all models)
  • ✅ Anthropic (Claude 3 family)

Coming soon:

  • 🔜 Google Gemini
  • 🔜 Cohere
  • 🔜 Mistral AI

Pricing

  • Free: 10,000 calls/month, 7-day retention
  • Pro: Unlimited calls, 90-day retention, $149/month
  • Enterprise: Custom retention, SSO, self-hosted options

View detailed pricing → 

Need Help?

Last updated on
Quick Start