vLLM Semantic Router 推出 Decision 2.0,单次前向回答多问题
vLLM 的 Semantic Router 出了 Decision 2.0,一次前向就能同时答多个问题还带各选项概率,做路由和分类的可以看看。
vLLM Semantic Router 团队发布了 Decision 2.0。新机制可以在一次前向传播(single forward pass)中处理同一输入下的多个问题,并输出每个选项的概率(per-option probabilities)。该能力适用于路由(routing)和分类(classification)场景。
Great work from the vLLM Semantic Router team on Decision 2.0. Multiple questions about one input, answered in a single forward pass with per-option probabilities. Useful for routing and classification.