{
  "product": {
    "name": "ziptoken",
    "tagline": "Less tokens. Same intelligence.",
    "description": "ziptoken is a REST API that compresses AI prompts and system messages by 30–65% before they reach LLM providers. It reduces token spend on OpenAI, Anthropic, Google, and other LLM APIs with a single API call and no changes to your stack.",
    "url": "https://ziptoken.ai",
    "category": "AI Infrastructure / LLM Cost Optimization",
    "launched": "2024"
  },
  "capabilities": [
    "Compress any text prompt or system message",
    "Model-agnostic: works with GPT-4, Claude, Gemini, Mistral, Llama, and any LLM",
    "Under 5ms p99 latency",
    "Up to 65% token reduction",
    "2.4 billion tokens processed per day",
    "Python and Node.js SDKs",
    "REST API with Bearer token auth"
  ],
  "api": {
    "base_url": "https://api.ziptoken.ai",
    "version": "v1",
    "primary_endpoint": {
      "method": "POST",
      "path": "/v1/compress",
      "auth": "Bearer sk-zip-...",
      "request_body": {
        "text": "string (required)",
        "model": "string (optional)"
      },
      "response": {
        "compressed_text": "string",
        "original_tokens": "number",
        "compressed_tokens": "number",
        "ratio": "number (0–1, lower is more compressed)",
        "tokens_saved": "number"
      }
    },
    "documentation": "https://ziptoken.ai/docs/api-reference"
  },
  "pricing": [
    {
      "plan": "Free",
      "price_usd_monthly": 0,
      "token_quota_monthly": 500000,
      "credit_card_required": false
    },
    {
      "plan": "Starter",
      "price_usd_monthly": 19,
      "token_quota_monthly": 10000000,
      "credit_card_required": true
    },
    {
      "plan": "Pro",
      "price_usd_monthly": 99,
      "token_quota_monthly": 100000000,
      "credit_card_required": true
    },
    {
      "plan": "Enterprise",
      "price_usd_monthly": null,
      "token_quota_monthly": null,
      "credit_card_required": true
    }
  ],
  "use_cases": [
    "Chatbots and virtual assistants with large system prompts",
    "RAG pipelines with verbose document context",
    "AI agents with long tool descriptions and conversation history",
    "Code generation with detailed instruction prompts",
    "Customer support automation",
    "Any application sending repetitive or verbose prompts to an LLM API"
  ],
  "links": {
    "home": "https://ziptoken.ai",
    "playground": "https://ziptoken.ai/playground",
    "quickstart": "https://ziptoken.ai/docs/quickstart",
    "api_reference": "https://ziptoken.ai/docs/api-reference",
    "pricing": "https://ziptoken.ai/pricing",
    "blog": "https://ziptoken.ai/blog",
    "status": "https://ziptoken.ai/status",
    "llms_txt": "https://ziptoken.ai/llms.txt",
    "llms_full": "https://ziptoken.ai/llms-full.txt"
  },
  "contact": {
    "support": "hello@ziptoken.ai",
    "twitter": "@ziptoken_ai"
  }
}