VIXOL — Intelligent LLM Cost Optimization

Save 40% on your LLM API costs with intelligent routing and automatic caching.

What is VIXOL?

VIXOL analyzes your queries and routes them intelligently:

• Simple queries → Groq (€0.0001) — fast and cheap
• Complex queries → Claude (€0.003) — high quality reasoning
• Repeated queries → Cache (€0.00) — instant free responses

Result: 40% cost savings on average, same quality.

Installation

npm install vixol

Quick Start

const VIXOL = require('vixol');

const vixol = new VIXOL({
  apiKey: 'your-api-key-here'
});

// Simple query → routes to Groq (cheap!)
const response = await vixol.query('What is the capital of France?');
console.log(response.result); // "Paris"
console.log(response.cost); // { price_eur: 0.0001, fee_eur: 0.000015 }

Features

✅ Intelligent Routing — Analyzes complexity, picks cheapest option ✅ Automatic Caching — Same query twice = free the second time ✅ Transparent Pricing — See exactly what you save ✅ Zero Configuration — Works out of the box ✅ Open Source — MIT licensed

API

query(prompt, options)

Send a query to VIXOL. Returns:

{
  result: "...",           // LLM response
  provider: "groq",        // Which LLM was used
  tokens: {
    in: 47,                // Input tokens
    out: 3                 // Output tokens
  },
  cost: {
    price_eur: 0.0001,     // What you pay
    fee_eur: 0.000015,     // VIXOL fee (15%)
    currency: "EUR"
  },
  balance_remaining: 99.99  // Your account balance
}

Options

• maxTokens (number) — Max response length (default: 1024)
• forceProvider (string) — Force specific LLM: "groq" or "anthropic"

Pricing

Free tier: Groq-only routing, unlimited queries Pro: 15% token fee, full smart routing Enterprise: Custom pricing, white-label, SLA

Get Started

1. Sign up: https://vixol.cloud
2. Create API key
3. npm install vixol
4. Start saving!

Support

• GitHub: https://github.com/thomasl-99/vixol
• Issues: https://github.com/thomasl-99/vixol/issues
• Email: [email protected]

License

MIT