A System One model is a decision model: it reads a state and answers typed questions — choose an option, score on a scale, or give the probability that a condition holds — in one forward pass, with calibrated confidence and no generated text.
A System Two model reasons by generating text token by token — in practice, a large language model. The term contrasts it with System One models, which return a typed decision in a single pass.
A decision model (in AI, an AI decision model) maps an input to an answer from a known set, with a probability for each answer, instead of producing open-ended text. System One models are decision models that take their questions at run time.
A System One model registry is a hub where System One models are published, versioned and compared, with the facts a decision model needs: the questions it answers, its calibration, latency and price. systemonemodels.ai is one — a Hugging Face built only for System One models.
A typed question is one question put to a System One model together with the shape of its answer: a choice among listed options, a score on a defined scale, or a noul probability. The type fixes what the model may return.
A choice question asks a System One model to pick exactly one option from a list you supply, and returns the chosen option with a probability for every option.
A score question asks a System One model to place the state on an ordered scale you define — for example 0 to 4, with a meaning for each level — and returns the score with a probability for every level.
Noul is the System One question type for yes/no conditions: it returns the probability that a condition described in the question holds for the state, as one number between 0 and 1.
The state is the input a System One model reads: the text, JSON, log or conversation that its typed questions are about. One state can be asked many questions in the same request.
A model’s confidence is calibrated when its probabilities match how often it is right: of all the answers it gives with 0.9 confidence, about 90% are correct. Calibration is what makes a threshold on the probability mean something.
Expected calibration error (ECE) measures how far a model’s confidence is from its accuracy: answers are grouped by confidence, and the gap between average confidence and accuracy in each group is averaged, weighted by group size. Lower is better; 0 is perfect.
Decision accuracy is the share of questions a System One model answers correctly on an evaluation set, reported as decision_accuracy — a number between 0 and 1 — in systemone.yaml.
Valid action rate is the share of a model’s answers that are allowed answers — an option that was offered, a level that is on the scale. A model that only scores the allowed answers reaches 100% by construction.
Decision latency is the time a System One model takes to answer a request, usually reported as the median (p50) and the 95th percentile (p95) in milliseconds, on stated hardware.
Confidence gating acts on a model’s answer only when its probability clears a threshold, and sends everything else to a fallback — a person, a larger model or a safe default. The threshold trades coverage for accuracy.
systemone.yaml is the manifest a System One model is published with: what it decides, its architecture and base model, licence, runtime, parameters, context length, evaluation numbers and, for hosted models, the provider and price.
NoulXP is the open standard for System One models: a model becomes a package that any engine runs without code written for it, proves it answers as the model’s own code does, and is asked the same way over HTTP or as an MCP tool.
A model tree is a model’s lineage: the base model it was fine-tuned, quantized or converted from, that model’s own base, and so on — together with the fine-tunes built on it.
An open-weights System One model publishes its checkpoint for anyone to download and run; a hosted-API model is available only as a paid endpoint from its maker. Some models are both.
A Jev alternative is a System One model other than TypeSafe AI’s Jev that answers the same kind of typed questions (choice, score and noul) with calibrated probabilities. Most are open-weights models you can run and fine-tune yourself, and several serve Jev’s own POST /v1/systemone API.
Laya is an open-weights System One model from Convai Innovations: a fully fine-tuned ModernBERT-large encoder with a decision head that scores one marker per option. It answers choice, score and noul questions in a single pass, under Apache-2.0.
LoRA (low-rank adaptation) fine-tunes a model by training small low-rank matrices added to its weights instead of the weights themselves. It makes adapting a System One model to your own decisions fast and cheap enough for a laptop.
