集体模型智能新论文：用 Model Eval、Mio、Mos 协调多模型协作
Toward Collective Model Intelligence: Characterizing and Coordinating Models
这篇论文做了个多模型调度系统，按任务自动选模型还能省钱，成本比单用最强模型低七成多，做路由和推理优化的可以看看。
arXiv 论文提出实现集体模型智能（Collective Model Intelligence）的三个组件：Model Eval 通过查询级性能估计与跨域预测构建模型能力画像，减少穷举评估；Mio（Models in One）按任务需求和成本做模型选择与协作；Mos（Model Serving OS）负责模型执行调度与共享部署。在九个公开基准上，Mio 1.0 Lite 取得最高平均分，单次请求成本比同分方法低 73.7%-97.5%，预热共享缓存使模型文件读取提速最高 23.3 倍。
Toward Collective Model Intelligence: Characterizing and Coordinating Models
Independently developed AI models accumulate diverse and complementary capabilities, but a growing model collection does not automatically realize Collective AI. Addressing this gap requires understanding model capabilities with limited evaluation, organizing complementary models to collaborate on diverse tasks, and executing model calls efficiently under resource constraints. We present a systematic approach to pursuing Collective Model Intelligence through three connected components. First, Model Eval constructs evidence-based capability profiles by combining query-level performance estimation with cross-domain prediction, reducing the need for exhaustive evaluation. Second, Mio (Models in One) organizes complementary model candidates and supports task-conditioned, cost-aware selection and collaboration as task requirements evolve. Third, Mos (Model Serving OS) separates model selection from execution and provides endpoint scheduling and shared model deployment and distribution. Experiments show lower capability-estimation errors under limited evaluation. Across nine public benchmarks, Mio 1.0 Lite achieves the highest average score and lowest average cost among compared methods, with per-request costs 73.7-97.5% lower than those of the best-scoring alternatives. A prewarmed shared cache accelerates model-file retrieval by up to 23.3 times. This work lays a foundation for Collective Model Intelligence by characterizing and coordinating the diverse capabilities of an expanding model universe.
