Cuanto AI

Client-facing AI API for Cuanto Labs. /v1/systemone is a drop-in, wire-compatible mirror of TypeSafe's Jev API — point the official TypeSafe SDK at this host and it just works. Everything else under /v1/* proxies to OpenRouter.

Endpoints

• POST /v1/systemone — TypeSafe-compatible Jev endpoint
• GET /v1/models — merged TypeSafe model list
• POST /v1/chat/completions, /v1/completions, /v1/embeddings, /v1/responses, /v1/messages, /v1/generation, /v1/rerank, /v1/images/*, /v1/audio/*, /v1/moderations — OpenRouter pass-through
• /v1/openrouter/* — direct OpenRouter pass-through (allowlisted)
• GET /health — liveness check

curl

curl https://<host>/v1/systemone \
  -H "Authorization: Bearer cl_live_<your key>" \
  -H "Content-Type: application/json" \
  -d '{
    "state": "Customer says their payment failed twice today.",
    "model": "jev-1.13.0",
    "questions": { "is_urgent": { "type": "noul", "instructions": "Is this urgent?" } }
  }'

TypeSafe SDK

import { TypeSafeClient } from "typesafe";

const client = new TypeSafeClient({
  apiKey: "cl_live_<your key>",
  baseURL: "https://<host>",
});