What is Decision 3.0? vLLM Semantic Router’s New Open Decision Models
vLLM Semantic Router drops Decision 3.0: five open-weight models (0.8B–27B) that answer choice, yes/no, and scoring questions on text and images without generating text. Direct classification without reasoning overhead.
Why it matters
A new open-weight model family optimized for structured decision tasks (classification, scoring) rather than generation. Relevant to practitioners building agent routing, content classification, and scoring pipelines who want Apache-2.0 alternatives to API-based classifiers.
The key facts
11 to knowvLLM Semantic Router released Decision 3.0
5 multimodal decision models
Model sizes: 0.8B to 27B parameters
Apache 2.0 license
Handles choice, yes/no, and score questions
Operates on text and images
Single-call inference, no text generation
5 open-weight multimodal models under Apache 2.0 license
Designed for choice, yes/no, and scoring questions on text and images
No generative text output (classification-only design)
Released October 2026
Go to the source
MarkTechPostmarktechpost.com
Publisher excerpt: vLLM Semantic Router released Decision 3.0, 5 Apache-2.0 multimodal decision models from 0.8B to 27B. They answer choice, yes/no and score questions about text and images in one call, without generating text.