How to use Gemini Decision 1 on Fikra API

Gemini Decision 1 is now available on Fikra API. It is a decision model, not a chat model: it returns calibrated answers for yes/no, choice and score questions, with zero output tokens.

The endpoint

The Decisions API is not OpenAI-compatible. Fikra exposes it at POST /v1/decisions with an API key. The model is fixed to gemini-decision-1.

curl https://api.fikraapi.co.ke/v1/decisions \
  -H "Authorization: Bearer $FIKRA_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "input": "The blue mug I ordered on Monday arrived cracked on Wednesday.",
    "questions": {
      "is_damaged": {
        "type": "noul",
        "instructions": "Did the item arrive damaged?"
      }
    }
  }'

Python

import requests
r = requests.post(
    "https://api.fikraapi.co.ke/v1/decisions",
    headers={"Authorization": "Bearer YOUR_FIKRA_KEY"},
    json={
        "input": "I was charged twice for my order.",
        "questions": {
            "department": {
                "type": "choice",
                "instructions": "Which department should handle this request?",
                "criteria": {
                    "billing": "Payments, invoices and refunds.",
                    "technical": "Problems using the product.",
                    "shipping": "Delivery and tracking.",
                    "other": "Anything else."
                }
            }
        }
    }
)
print(r.json()["answers"]["department"])

Question types

noul returns a yes probability between 0.01 and 0.99. choice returns the top option and probabilities. score returns a probability-weighted level from an ordered rubric. Every answer includes a confidence score. Use confidence thresholds before acting on a decision.

Pricing

$0.12 per 1M input tokens. Output is 0 because the model returns no text tokens. Media input is supported by the same Decisions API schema.

See the Decisions API guide for the full request reference.