vLLM Semantic Router team releases Decision 2.0, six open decision models from 0.6B to 27B

The vLLM Semantic Router team has released Decision 2.0, a set of six open decision models ranging from 0.6B to 27B parameters, according to a vLLM post on X. The models answer multiple questions about one input in a single forward pass and output per-option probabilities, which the post presents as useful for routing and classification.

A quoted post from Xunzhuo Liu claims each of the six models tops the same-size open models it was compared against on the Jev Decision Index 0.3. This benchmark result is reported as a claim by Liu; no independent evaluation or detailed results appear in the available reports.