English(EN) Together Compiler: Continual Compilation for AI Inference

Together AI发布用于持续AI推理优化的编译器

作者 PulseAugur 编辑部 · [1 个来源] · 2026-11-14 05:00

Together AI的研究人员开发了Together Compiler (TKCC)，这是一个专为持续优化AI推理工作负载而设计的新颖系统。TKCC通过采用基于代理的循环来平衡测量吞吐量和计算保真度，从而应对快速发展的AI模型和多样化流量模式的挑战。该系统引入了分层持久引擎以减少引擎启动开销，并使用成本模型来选择最佳并行配置。在生产环境中，TKCC已展示出显著的性能提升，在各种前沿模型和硬件平台上超越了高度优化的内核和开源引擎，为其舰队服务了数万亿个token。 AI

影响该编译器可以显著提高在生产环境中部署和运行AI模型的效率和速度。

排序理由该条目是一篇研究论文，详细介绍了一个用于AI推理优化的新系统。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.AI 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 16 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该条目是一篇研究论文，详细介绍了一个用于AI推理优化的新系统。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: infra, model release; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.AI TIER_1 English(EN) · Reyna Abhyankar, Dylan Lim, Drew Wadsworth, Venkata Hasith Vattikuti, YuTong Julia Zhang, Jung Hwan Heo, Taeksang Kim, Kesavan Ramakrishnan, Evangelene Stanley, Xinyu Fang, Daniel Y. Fu, Simran Arora · 2026-11-14 05:00

Together Compiler：AI推理的持续编译

arXiv:2611.14681v1 Announce Type: new Abstract: Today's AI workloads are rapidly changing: new models are released more often, model architectures are increasingly diverse, and traffic distributions change frequently. The challenge we face at Together AI is continual optimization…

本文如何被排名

报道来源 [1]

Together Compiler：AI推理的持续编译

相关实体

相关话题