ziptoken Documentation

ziptoken compresses AI prompts before you send them to any LLM — reducing token usage by 25–55% with no measurable quality loss.

Base URL

https://api.ziptoken.ai

How it works

  1. 1

    Obtain an API key

    Create a free account and generate an API key from your dashboard.

  2. 2

    POST to /api/v1/compress

    Send your prompt text. Receive a compressed version — typically 25–55% shorter.

  3. 3

    Forward to your LLM

    Use the compressed text with Claude, GPT-4, Gemini, or any LLM. Same quality, fewer tokens.

Authentication

All API requests must include your key in the Authorization header:

Authorization: Bearer zt_live_xxxxxxxxxxxx

Example request

curl -X POST https://api.ziptoken.ai/api/v1/compress
  -H "Content-Type: application/json"
  -H "Authorization: Bearer zt_live_xxxx"
  -d '{"text": "Your prompt here", "mode": "balanced"}'
Response
{
  "data": {
    "compressed":       "Your prompt here",
    "originalTokens":    312,
    "compressedTokens":  134,
    "ratio":            0.57,
    "qualityScore":       4.8
  }
}
Live

API Tester

Test the API directly from your browser — no setup needed.

Try it — POST /api/v1/compress

Rate limits

PlanTokens / monthRate limit
Free50,00060 req / min
Starter2,000,000300 req / min
Pro10,000,0001,000 req / min
EnterpriseUnlimitedCustom