AWS releases open-source model for choosing AI workflow actions

Amazon Web Services has released Strands Decider 2B, an open-source model that selects among predefined options instead of generating open-ended text. It can choose an AI agent's next tool, route customer requests, score outputs and return confidence scores. Built on Qwen3.5-2B-Base, the model replaces the language-generation component with a layer that scores available choices. AWS says it can run on Apple silicon Macs and CPUs; benchmark testing recorded median response times of around 115 milliseconds on an RTX 3090. The release through AWS's experimental Strands Labs initiative targets routine decisions in automated workflows, where developers may not need a larger model's text-generation capabilities.