Quickstart

Prerequisites

1. Create an account at trytresor.com.
2. Go to the API Dashboard and create an API key. Copy it — it is only shown once.

Make your first request

Replace YOUR_API_KEY below and run:

curl https://api.trytresor.com/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-oss-120b",
    "messages": [
      {"role": "user", "content": "Hello!"}
    ],
    "stream": true
  }'

You should see the model's response streamed back as Server-Sent Events.

How it works

1. Your request is authenticated using your API key.
2. The router selects the best available provider and forwards the request to a confidential enclave.
3. The response is streamed back via SSE — neither Tresor nor any third party can read your data.

This is not a proxy. The inference runs inside a hardware-isolated enclave with encrypted memory. Learn more about zero-access.

Next steps

• Authentication — API key format, rate limits, and security.
• SDK examples — Python, Node.js, and Go.
• Models — see available models and pricing.