Decision from text

Classify messages, transcripts, records, and JSON with HeyGen Decision 1: put the evidence in state and ask named questions about it.

Text is the base of every decision. Put the material to judge in state and the judgment you want in each question’s instructions. A text-only request omits assets.

state accepts a string, a JSON object, or a JSON array. Objects and arrays are serialized as text and read as one shared state, not as a batch of separate requests.

​

Label related material with an object

A string is enough for a single message. Use an object when the decision depends on several pieces, such as a customer message and the policy it is checked against:

curl -X POST https://api.heygen.com/v3/models/decisions \
  -H "X-Api-Key: $HEYGEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "heygen-decision-1",
    "state": {
      "message": "Please cancel the extra seat added yesterday.",
      "policy": "Unused seats added within the last seven days can be refunded.",
      "seat_used": false,
      "days_since_purchase": 1
    },
    "questions": {
      "refund_eligible": {
        "type": "noul",
        "instructions": "Does the request meet the refund policy?",
        "criteria": {
          "true": "All requirements for a refund are met by the supplied evidence.",
          "false": "At least one requirement is not met or lacks supporting evidence."
        }
      },
      "intent": {
        "type": "choice",
        "instructions": "What does the customer want?",
        "criteria": {
          "cancel_seat": "Remove a seat or license",
          "cancel_plan": "End the whole subscription",
          "other": "Anything else"
        }
      }
    }
  }'

Answers come back in data.answers under the same names. Question types covers each answer’s shape.

​

Write a state the model can judge

• Include the content, not a link. URLs in state are read as text; paste the document, page, or transcript you want evaluated.
• Name the parts. Keys such as message, policy, and history tell the model what each piece is.
• Keep one decision per request. Questions in a request share the same state and run independently. When a question depends on an earlier answer, ask it in a later request.
• Watch the context budget. heygen-decision-1 reads up to 128k tokens per model read, and heygen-decision-1-turbo (preview) up to 256k. See Context limits.

For spoken content, put the transcript in state.