vLLM releases open Decision 2.0 models: 64 questions in one pass at 63ms
vllm_project · x · 2026-10-03
The vLLM Semantic Router team released Decision 2.0, Apache-2.0 open decision models spanning 0.6B to 27B that load with Transformers. The 0.6B/0.8B/2B/4B variants rank #1 at their size on the Jev Decision Index (27B is #3 overall), scoring up to 2.5× Decision 1.0 at equal size. One forward pass answers 64 questions about a request in 63ms on a single GPU—roughly 20× faster than the Jev API per decision. Built by Xunzhuo Liu's team for vLLM Semantic Router.
Related event: vLLM Team Open-Sources Decision 2.0 Decision Models from 0.6B to 27B(3 posts)→
More from Models
- Mystery Model 'Gestalt' Tops New Frontier Design Benchmark by 214 Points — soleio · 2026-11-19
- mrfakename releases abliterated GLM-5.3-Flash and Qwen3.8-Flash-Next — mrfakename0 · 2026-11-19
- Physician: AI competition now measured in weeks, GPT-6 versions 7 days apart — ramos_casals · 2026-11-19
- micro1's flow qc model screens training data for benchmark leaks before release — Exp_Mark · 2026-11-19
- Leaker predicts new DeepSeek paper this week, scheduled for November 11 — teortaxesTex · 2026-11-19
- Gemini 4 Neon rollout begins: leaked UI shows 3-tier Gemini 4 lineup with Carbon locked — testingcatalog · 2026-11-19