AWS’s Strands Agents team has announced the release of Strands Decider 2B, an open-source decision model designed to enhance decision-making processes. Built on the Qwen3.5-2B-Base architecture, this model is capable of delivering choices, yes/no probabilities, and scores with calibrated confidence in a single forward pass, without generating text.
Strands Decider 2B operates with a median response time of just 115 milliseconds on an RTX 3090, making it a highly efficient option for applications requiring rapid decision-making. The model has achieved a score of 0.723 on the JevBench public set, positioning it as a competitive local solution for routing and tool selection tasks.
This release underscores AWS’s commitment to advancing AI capabilities, providing developers and organizations with powerful tools to streamline decision-making processes in various domains.
Compiled automatically by the Tech AI Newsdesk from public AI-news sources and summarised in our own words.
