AITOL — AI Token Optimization Layer

Reduce your LLM API costs by 40–66% with one command.

AITOL is a local proxy that sits between your AI tools and the LLM API. It compresses prompts, caches semantic duplicates, and routes to the cheapest capable model — automatically.

Install

npm install -g aitol

Quick Start

aitol init        # enter your API keys
aitol start       # start proxy on localhost:4141
aitol status      # see live savings

Connect Your Tools

aitol plugin add cursor       # Cursor
aitol plugin add claude-code  # Claude Code
aitol plugin add vscode       # VS Code
aitol plugin add chrome       # ChatGPT / Claude.ai / Gemini

How It Works

• Semantic Cache — returns cached responses for similar prompts (66%+ hit rate)
• MRL Compression — compresses natural language prompts by 45–56%
• CCE Compression — compresses code-heavy prompts by 34–44%
• Smart Routing — picks the cheapest model that fits the task

Proxy Endpoints

Point any OpenAI or Anthropic SDK at AITOL:

// OpenAI SDK
const openai = new OpenAI({
  apiKey: process.env.OPENAI_API_KEY,
  baseURL: "http://localhost:4141/openai"
});

// Anthropic SDK
const anthropic = new Anthropic({
  apiKey: process.env.ANTHROPIC_API_KEY,
  baseURL: "http://localhost:4141/anthropic"
});

Links

• Website: https://aitol.dev
• GitHub: https://github.com/Anant2507/AITOL
• Docs: https://aitol.dev/docs