Model overview
What this model is for
Decision-2.0-Lux-9B is a text-only 9B model designed for typed decisions rather than conversational prose. It answers caller-defined choice, yes/no, and score questions by returning probability distributions, making it potentially useful for routing, classification, policy selection, structured evaluation, and other applications that need machine-readable decision outputs. Its listed use case is typed decisions, and the supplied record does not identify additional capabilities such as coding, vision, tool calling, or unrestricted generation. The model is described as part of vLLM Semantic Router’s Decision 2.0 family and is licensed under Apache-2.0, but developers should review the Hugging Face model card and repository contents for implementation details, output schemas, training information, and any usage conditions. The context window, inference speed, and pricing are not disclosed in the supplied material. This model is therefore best considered when calibrated probability distributions are more useful than generated explanations or prose responses. The Hugging Face page is the listed official source; no publication or update date is disclosed, so freshness of the technical documentation should be verified before deployment.