Amanai Docs

System One

Jev 1.13 returns typed decisions and probabilities from a state and a set of questions.

TypeSafe AI
TypeSafe AI · Jev 1.13
Native structured evaluation · amanai/jev-1.13

Availability

Use POST /v1/systemone with model amanai/jev-1.13 and an Amanai API key. The vendor is TypeSafe AI. System One is a separate API surface, not a chat, embeddings, rerank, or media model.

Use the native contract. Send state and questions; do not send chat messages or expect a generated text completion.

Native contract

A request pairs a state (text or structured JSON object) with named questions. Each question declares its output type and an instruction. The response preserves question IDs under answers.

Question typeCriteriaAnswer
noulOptional descriptions for true and false.A noul value from 0 to 1.
choice2–64 named options.Selected option, confidence, and option probabilities.
score2–64 ordered labels.Numeric score, confidence, and label probabilities.

Send 1–64 questions per request. Question IDs are stable keys in the response.

Request example

Authenticate with a Bearer API key and send JSON to POST /v1/systemone.

bash
curl https://api.amanai.dev/v1/systemone \
  -H "Authorization: Bearer $AMANAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "amanai/jev-1.13",
  "state": "My payments have failed for three days.",
  "questions": {
    "urgent": {
      "type": "noul",
      "instructions": "Is this urgent?"
    },
    "team": {
      "type": "choice",
      "instructions": "Which team should handle it?",
      "criteria": {
        "billing": "Payment issues",
        "technical": "Product bugs"
      }
    },
    "frustration": {
      "type": "score",
      "instructions": "How frustrated is the customer?",
      "criteria": ["Calm", "Frustrated", "Angry"]
    }
  }
}'

Response example

Example response; probabilities and token counts vary with the request.

bash
{
  "model": "amanai/jev-1.13",
  "answers": {
    "urgent": { "type": "noul", "noul": 0.96 },
    "team": {
      "type": "choice",
      "choice": "billing",
      "confidence": 0.9,
      "probabilities": { "billing": 0.94, "technical": 0.06 }
    },
    "frustration": {
      "type": "score",
      "score": 1.2,
      "confidence": 0.7,
      "legend": { "0": "Calm", "1": "Frustrated", "2": "Angry" },
      "probabilities": { "0": 0, "1": 0.8, "2": 0.2 }
    }
  },
  "usage": { "input_tokens": 100, "output_tokens": 30 }
}

usage records input and output tokens even though this model has no customer charge.

Billing

amanai/jev-1.13 costs 0 credits per request while the offer is active. Currently free for a limited time while the upstream free offering remains available. Free access is not guaranteed and future requests may be priced differently. Token usage is still reported. See Billing & pricing for the System One rate.