亚马逊云服务(AWS)推出了名为Strands Decider 2B的开源决策模型1。该模型基于Qen3.5-2B构建，专注于在预定选项之间进行排序并提供置信度评分，而非生成文本内容1。作为一款高速、低成本的工具，Strands Decider 2B足够轻量，可以在本地运行1，旨在满足AWS客户对更轻量级人工智能工具的需求。
该模型由亚马逊杰出工程师Marc Brooker开发，创意来自TypeSafe公司的Jev模型1。Strands Decider 2B已完全开源并可供使用1。TypeSafe首席执行官Diogo Almeida对此表示："他们可能低估了让这些模型真正聪明的难度"1。此外，数十个类似的决策模型已由研究人员开发，显示出该领域的广泛兴趣1。
Amazon Web Services has launched Strands Decider 2B, an open-source decision-making model designed to rank among predefined options and provide confidence scores rather than generate text 1. Built on Qen3.5-2B, the lightweight model is small enough to run locally, addressing demand from AWS customers for more compact AI tools that can operate without cloud infrastructure 1.
The model was developed by Amazon distinguished engineer Marc Brooker and is inspired by TypeSafe's Jev model 1. According to TypeSafe CEO Diogo Almeida, developers may be "underestimating the difficulty in making these models truly intelligent" 1. The release comes as dozens of similar decision models have emerged from researchers across the field, reflecting widespread interest in specialized AI tools designed for specific decision-making tasks rather than general text generation 1.
评论
还没有评论，欢迎留下第一条。