# Decision Models Return Calibrated Probabilities Without Generating a Single Token

Liquid AI's d1 and a new Ollama endpoint answer typed yes-or-no, choice and score questions in one call, with a 9-billion-parameter open model averaging 91 milliseconds per decision on a laptop.

- Published: 2026-09-30T06:34:52.927Z
- Canonical: https://polylog.news/ai/2026-09-30/decision-models-return-calibrated-probabilities-without-gene
- Publisher: Polylog (AI desk)
- Section: tech
- Sources: [Polylog editors](https://polylog.news), [Ollama](https://ollama.com/blog/ollama-now-supports-jev-style-decision-models), [arXiv cs.LG](https://arxiv.org/abs/2609.31631), [arXiv cs.AI](https://arxiv.org/abs/2609.35833)

Liquid AI has released d1, the first of what it calls decision models. The model takes text or JSON context and returns calibrated probabilities over a predefined set of answers in a single call, producing zero generated tokens. Three respo…

This story is for subscribers. Read it in full at https://polylog.news/ai/2026-09-30/decision-models-return-calibrated-probabilities-without-gene (subscription information: https://polylog.news/pricing).