Jev API · Jev AI

Jev API

Call the Jev model from your own code. The Jev AI endpoint takes the same JSON as TypeSafe's System One API — a state, a model and a map of typed questions — and returns one typed answer per question.

Your Jev API keys

Endpoint

POST https://jev-ai.pro/api/v1/systemone
Authorization: Bearer <JEV_AI_API_KEY>
Content-Type: application/json

Request

curl https://jev-ai.pro/api/v1/systemone \
  -H "Authorization: Bearer $JEV_AI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "state": "Help! My payouts have been failing for 3 days.",
    "model": "jev-latest",
    "questions": {
      "is_urgent": { "type": "noul", "instructions": "Does this convey urgency?" },
      "department": {
        "type": "choice",
        "instructions": "Which team should handle this?",
        "criteria": { "billing": "Payments and refunds", "technical": "Bugs and outages", "sales": "Pricing" }
      },
      "frustration": {
        "type": "score",
        "instructions": "How frustrated is the customer?",
        "criteria": ["Calm", "Frustrated", "Very angry"]
      }
    }
  }'

Response

{
  "model": "jev-1.13.0",
  "answers": {
    "is_urgent": { "type": "noul", "noul": 0.95 },
    "department": {
      "type": "choice", "choice": "billing", "confidence": 0.98,
      "probabilities": { "billing": 0.99, "technical": 0.01, "sales": 0.0 }
    },
    "frustration": {
      "type": "score", "score": 1.04, "confidence": 0.94,
      "legend": { "0": "Calm", "1": "Frustrated", "2": "Very angry" },
      "probabilities": { "0": 0.0, "1": 0.96, "2": 0.04 }
    }
  },
  "usage": { "input_tokens": 379, "output_tokens": 70 }
}

Every answer comes back under the question id you chose. Choice and score answers include probabilities and a confidence between 0 and 1; a noul answer is the probability of yes.

Official SDK

// The official TypeSafe JavaScript SDK works unchanged: point it at Jev AI.
import { TypeSafeClient, choice } from '@typesafe-ai/sdk'

const client = new TypeSafeClient({
  apiKey: process.env.JEV_AI_API_KEY,
  baseURL: 'https://jev-ai.pro/api',
})

const response = await client.systemOne({
  state: { document: 'I was charged twice. Please fix this ASAP.' },
  questions: {
    category: choice('What is this ticket about?', { billing: null, technical: null, other: null }),
  },
})
console.log(response.answers.category.choice) // "billing"

Tested with @typesafe-ai/sdk 0.6.0. GET /api/v1/models lists the model names and GET /api/v1/credits returns your balance.

Models, limits and billing

Models	`jev-latest, jev-preview, jev-1.13.0` (aliases currently resolve to `jev-1.13.0`)
Question types	`noul` (yes/no), `choice`, `score` · up to 64 per request
Input	Text only: a string, JSON object or array · request body up to 256 KB · model context 64k tokens
Price	1 credit per 1,000,000 input tokens as reported by Jev; output tokens are free. Partial blocks carry over to the next request.
Balance	A request needs at least 1 credit. Response headers `X-Jev-Credits-Charged` and `X-Jev-Credits-Remaining` show the metering.
Rate limit	300 requests per minute per account; `429` with `Retry-After` when Jev is busy.
Errors	`401` bad key · `402` no credits · `422` invalid request · `429` slow down · `502/504` upstream failure (not billed)
Privacy	API requests are not stored: we keep token counts and status for billing, never your state or answers.

Jev AI is an independent service and is not affiliated with or endorsed by TypeSafe. Jev answers are probabilistic; review important decisions before acting on them.