SysOne
Typed decision models running entirely in your browser. Text stays local; model files are downloaded and cached by the browser.
Model
State
Questions
Assistant
Describe what you want to decide. The assistant sets the questions up, loads the decision model selected above (or the one you name in the chat), runs it, and explains what came back.
Voice
The same conversation as the Assistant, out loud: say what you want to decide, it sets the questions up, runs the decision model and reads the answer back. The speech models are separate from the decision model and are loaded on demand.
Fixed task
The same two questions for every model — “Which team should handle this?” (billing / technical support / sales) and “Is the customer angry?” — so runs are comparable across models.