Skip to content
Decisio

DocsStart

4 min read · updated

Your first decision

One request, three questions and the answer to each, taken apart line by line.

A decision is a text and typed questions about it. You send them in one request, and the server answers each question on its own: a probability for every option. This page takes one recorded decision apart; hover or focus a line of the figure to see its partner.

Figure · The request and its response

Every question gets its own answer, with a probability for every option.

Request

{
  "state": "I was charged twice for October. Both charges are 49.00 EUR, same day. Please refund one.",
  "questions": {
    "urgent": {
      "type": "noul",
      "instructions": "Does this ticket need a response within the hour?"
    },
    "queue": {
      "type": "choice",
      "instructions": "Which team should handle this ticket?",
      "criteria": {
        "billing": "Invoices, payments, refunds, prices, VAT",
        "access": "Login, passwords, two-factor, SSO, permissions, invitations, account security",
        "bug": "The product behaves wrongly: errors, crashes, wrong numbers, lost data",
        "feature": "A request for something the product does not do yet",
        "cancellation": "Cancelling, pausing or closing an account, deleting data",
        "other": "Press, partners, jobs, sales pitches, wrong address, thanks, general questions"
      }
    },
    "impact": {
      "type": "score",
      "instructions": "Rate the business impact using only the reported facts.",
      "criteria": [
        "No function impaired",
        "One user impaired, with a workaround",
        "Many users blocked from a core function",
        "Data loss or legal exposure"
      ]
    }
  }
}

Response (abridged)

{
  "answers": {
    "urgent": {
      "type": "noul",
      "noul": 0.159
    },
    "queue": {
      "type": "choice",
      "choice": "billing",
      "probabilities": {
        "billing": 0.962,
        "access": 0.011,
        "bug": 0.016,
        "feature": 0.004,
        "cancellation": 0.005,
        "other": 0.003
      }
    },
    "impact": {
      "type": "score",
      "probabilities": {
        "0": 0.256,
        "1": 0.648,
        "2": 0.053,
        "3": 0.042
      }
    }
  }
}

Hover or focus a line: its partner on the other side lights up, and what it is appears here.

Recorded decision: Gemma 4 31B, decisio 0.9.0, RTX PRO 6000 at 600 W, 2026-10-08; record 1 of 60 of decisio-examples' support-routing run; the response is abridged and its probabilities are rounded to three places.

TL;DR

Section titled “TL;DR”

With a server running (Get started starts one), send the request and read the answer:

curl http://127.0.0.1:8000/v1/systemone \
  -H 'Content-Type: application/json' \
  -d '{
  "state": "I was charged twice for October. Both charges are 49.00 EUR, same day. Please refund one.",
  "questions": {
    "urgent": {
      "type": "noul",
      "instructions": "Does this ticket need a response within the hour?"
    },
    "queue": {
      "type": "choice",
      "instructions": "Which team should handle this ticket?",
      "criteria": {
        "billing": "Invoices, payments, refunds, prices, VAT",
        "access": "Login, passwords, two-factor, SSO, permissions, invitations, account security",
        "bug": "The product behaves wrongly: errors, crashes, wrong numbers, lost data",
        "feature": "A request for something the product does not do yet",
        "cancellation": "Cancelling, pausing or closing an account, deleting data",
        "other": "Press, partners, jobs, sales pitches, wrong address, thanks, general questions"
      }
    },
    "impact": {
      "type": "score",
      "instructions": "Rate the business impact using only the reported facts.",
      "criteria": [
        "No function impaired",
        "One user impaired, with a workaround",
        "Many users blocked from a core function",
        "Data loss or legal exposure"
      ]
    }
  }
}'

Run against the CPU stand-in at v0.11.1 on 2026-10-14: the answers are the stand-in's, not the measured system's.

{
  "answers": {
    "urgent": {
      "type": "noul",
      "noul": 0.159
    },
    "queue": {
      "type": "choice",
      "choice": "billing",
      "probabilities": {
        "billing": 0.962,
        "access": 0.011,
        "bug": 0.016,
        "feature": 0.004,
        "cancellation": 0.005,
        "other": 0.003
      }
    },
    "impact": {
      "type": "score",
      "probabilities": {
        "0": 0.256,
        "1": 0.648,
        "2": 0.053,
        "3": 0.042
      }
    }
  }
}

Recorded output.

Recorded decision: Gemma 4 31B, decisio 0.9.0, RTX PRO 6000 at 600 W, 2026-10-08; record 1 of 60 of decisio-examples' support-routing run; the response is abridged and its probabilities are rounded to three places.

WordMeaning
noula yes/no question; its answer is the probability of yes
choicea choice among named options (up to 255); the answer names the likeliest and gives every option’s probability
scorea level on an ordered scale, lowest first; the answer is a probability for each level
criteriawhat the options are: for a choice, option keys and their descriptions; for a score, the levels’ descriptions

1. The text

Section titled “1. The text”

state is the text the questions are about. A prompt, the state with one question, can hold up to 32,768 tokens, and the server reads the state once for all the questions in the request.

2. The questions

Section titled “2. The questions”

Each question has a name you choose, which is also its key in the answer, a type and instructions. The options of a choice are its criteria: a key for each, with a description the model reads. Leave instructions out and a fixed default per type is used.

3. The answer

Section titled “3. The answer”

The response has one answer per question, under the same name. A choice answers with its likeliest option and the probability of every option, so you can read how sure it is as well as what it picked.

4. One question on its own

Section titled “4. One question on its own”

Each question is scored on its own, so a question in a request is answered as if it had been sent alone. On the Qwen base its probabilities are the same; on the Gemma bases the choice is the same and the probabilities agree to within 0.035 (12B) and 0.0073 (31B). Ask only the queue:

curl http://127.0.0.1:8000/v1/systemone \
  -H 'Content-Type: application/json' \
  -d '{
  "state": "I was charged twice for October. Both charges are 49.00 EUR, same day. Please refund one.",
  "questions": {
    "queue": {
      "type": "choice",
      "instructions": "Which team should handle this ticket?",
      "criteria": {
        "billing": "Invoices, payments, refunds, prices, VAT",
        "access": "Login, passwords, two-factor, SSO, permissions, invitations, account security",
        "bug": "The product behaves wrongly: errors, crashes, wrong numbers, lost data",
        "feature": "A request for something the product does not do yet",
        "cancellation": "Cancelling, pausing or closing an account, deleting data",
        "other": "Press, partners, jobs, sales pitches, wrong address, thanks, general questions"
      }
    }
  }
}'

Run against the CPU stand-in at v0.11.1 on 2026-10-14: the one question alone, as the stand-in answers it.

What next

Section titled “What next”