API

An OpenAI-compatible endpoint backed by the same models and credits as the app. Keys are hashed on our side and shown to you once.

Documentation

Base URL https://tacit-lac.vercel.app/api/v1. Authenticate with Authorization: Bearer veil_sk_….

POST /chat/completions — model, messages, stream, max_tokens, temperature. Images as base64 data: URLs only.
GET /models — models configured on this deployment.
Extension: "privacy": "smart" | "strict" | "off" runs the filter on our server and restores placeholders in the response. Default: off. Unlike the app, your raw text reaches our server in this mode (it is not stored or logged).
Rate limit: 120 requests/minute per key. Responses include x-ratelimit-* headers.
Billing: credits per token at each model's rate; usage.credits is returned with every response. Requests fail with 402 when the balance is empty.

curl https://tacit-lac.vercel.app/api/v1/chat/completions \
  -H "Authorization: Bearer $VEIL_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "auto",
    "stream": true,
    "privacy": "smart",
    "messages": [{"role": "user", "content": "Summarise the attached notes for Priya."}]
  }'