Liquid AI releases decision model D1 (docs.liquid.ai)

Liquid AI has launched its new decision model, D1, which introduces three distinct query types—Noul, Choice, and Score—designed to enhance decision-making processes through probabilistic outputs. The Noul type answers yes/no questions with a probability score, the Choice type provides a probability distribution for selecting from multiple options, and the Score type allows for rating on an ordered scale while accounting for uncertainty. These primitives empower developers to implement more nuanced and weighted decision-making within their applications. The release of D1 is significant for the AI/ML community as it streamlines the integration of complex decision frameworks into various applications, from customer support systems to resource allocation. By allowing users to generate multiple types of outputs from a single API call, D1 reduces the complexity and overhead traditionally associated with retrieving structured insights from models. This feature enables more efficient processing of contextual data—whether in text or structured formats—while delivering comprehensive evaluations that consider classification, routing, and prioritization together, ultimately fostering more robust AI-driven solutions.