# meraGPT: state-decider-1

> meraGPT's hosted decision model (state-decider-1). Answers yes/no, choice and rubric-score questions over one state as calibrated distributions in a single pass, on the System One schema, so the typesafe-sdk works by changing its base URL.

- Page: https://systemonemodels.tech/meragpt/state-decider-1
- API: https://api.systemonemodels.tech/v1/models/meragpt/state-decider-1

## Facts

| | |
|---|---|
| Maker | meraGPT (https://systemonemodels.tech/meragpt) |
| Decides | choice, score, noul, classify, route |
| Architecture | state-decider |
| Context | 4K tokens |
| Licence | proprietary |
| Availability | Hosted API |
| Released | 2026-09-22 |
| Latest version | 2026.09 |

## Hosted API

- Provider: meraGPT (https://meragpt.com/models/state-decider-1)
- Docs: https://meragpt.com/docs
- Input: $0.030/MTok · Output: Free

## Reported evaluation

Suite: typed-decisions benchmark (400 cases, 2,000 decisions; teacher-ensemble references), single run. Numbers are the publisher's own.

- Decision accuracy: 76.8%

## Model card

<!-- generated by scripts/seed_catalog.py; edit content/models/catalog.yaml -->

# Decider 1

meraGPT's hosted decision model (state-decider-1). Answers yes/no, choice and rubric-score questions over one state as calibrated distributions in a single pass, on the System One schema, so the typesafe-sdk works by changing its base URL.

Not related to Mapika's open decider models. Text only; a request holds at most 4,096 tokens and a Choice question at most ten options. On a 400-case, 2,000-decision typed-decisions benchmark whose reference answers come from a teacher-model ensemble, meraGPT reports 0.768 accuracy (Jev 1.13: 0.727), KL 0.096 from the references (Jev: 1.442) and Brier 0.052 (Jev: 0.148), from a single run. $0.03 per million input tokens, output free, with $1 of free credit. Weights are proprietary.

## What it decides

- **choice** — picks one option from a set
- **score** — places the input on an ordered scale
- **noul** — answers a yes/no question with one calibrated probability
- **classify** — assigns a category from a fixed taxonomy
- **route** — sends the input to one of several destinations

## At a glance

| | |
|---|---|
| Maker | meraGPT |
| Released | 2026-09-22 |
| License | proprietary |
| Reported accuracy | 76.8% |

## Hosted API

Served by **meraGPT** — $0.03/MTok input, $0/MTok output. [Get access](https://meragpt.com/models/state-decider-1) · [API docs](https://meragpt.com/docs).

## Read more

- [Model page](https://meragpt.com/models/state-decider-1)
- [Launch post](https://meragpt.com/blog/introducing-decider-1)

---

*This page was opened by System One for meraGPT, who can claim the organisation and take it over at any time.*

---

From System One Models — https://systemonemodels.tech/ · every System One model: https://systemonemodels.tech/system-one-models