---
格式版本: 2
标题: "Cirrascale supports AMD GPUs"
原文链接: "https://www.lightreading.com/ai-machine-learning/cirrascale-supports-amd-gpus"
发布日期: "2026-07-14"
发布时间校准状态: "found"
发布时间来源: "llm:strict_original_body"
发布时间证据: "div class=Contributors-InfoWrapper: July 14, 2026"
发布时间校准原因: "该日期位于 Contributors-InfoWrapper 区域，紧邻标题，符合文章发布时间的典型位置特征，且与当前 YAML 发布日期一致。"
发布时间校准置信度: "100"
发布时间候选数量: 39
发布时间严格候选数量: 14
发布时间原页读取状态: "原页面来自已抓取 HTML"
发布时间未找到原因: "候选日期无效或 LLM 未确认"
发布时间校准时间: "2026-07-17T15:14:03+08:00"
发现时间: "2026-07-17T15:03:54+08:00"
入库时间: "2026-07-17T07:19:57.724Z"
来源平台: "Tavily（AgentKey）"
搜索渠道: "tavily_web"
搜索词: "Rack level server"
匹配关键词:
  - "GPU"
  - "UALink"
相关厂家:
  - "AMD"
相关专家:
  []
内容类型: "网页"
抓取工具: "AgentKey Scrape"
清洗工具: "AgentKey Markdown + LLM 正文裁剪"
原始附件:
  []
图片摘要:
  - "✗ ./assets/img-1168dd5c.webp | photo | 图片Alt文本指向另一篇文章（关于管理服务和技术债务），与正文Cirrascale支持AMD GPU的主题无关，属于网页侧边栏或推荐阅读配图。"
  - "✗ ./assets/img-553aff43.webp | photo | 图片Alt文本指向另一篇文章（关于2025年人员缩减），内容为历史照片，与正文AI基础设施主题完全无关。"
  - "✗ ./assets/img-a9ab09ab.webp | photo | 图片Alt文本指向另一篇文章（关于Nokia和AI网络），展示的是Nokia展台，正文未提及Nokia，属于无关配图。"
  - "✗ ./assets/img-c4e28f47.webp | photo | 图片Alt文本指向另一篇文章（关于光纤宽带客户AI使用调查），虽然涉及AI但主题不符，正文未提及该调查，属于无关配图。"
AI优质: "是"
AI打分: 95
AI分档: "高置信优质"
AI质检状态: "通过"
AI打分理由: "深度介绍AMD Helios 72-GPU rack-scale平台，包含compute tray、HBM4、UALoE互连等细节，并宣布Cirrascale云服务部署，信息新颖且完整。"
AI质检模型: "deepseek-v4-pro"
AI质检时间: "2026-07-30T01:07:11+08:00"
AI主题相关性: 20
AI来源权威性: 10
AI新颖性: 20
AI技术细节: 20
AI商业部署信号: 15
AI完整性: 10
采集批次: "2026年7月17日15点03分52秒"
采集批次ID: "20260717-150352-345"
去重键: "https://www.lightreading.com/ai-machine-learning/cirrascale-supports-amd-gpus"
---

Cirrascale brings AMD Instinct™ MI400 Series GPUs to its AI innovation cloud on open, standards-based infrastructure. #pressrelease

Cirrascale Cloud Services, the expert neocloud built for Private AI, today announced support for the AMD Helios rackscale solution and AMD Instinct™ MI400 Series GPUs across its AI innovation cloud. The addition brings AMD rack-scale AI infrastructure, powered by AMD Instinct™ MI455X GPUs, to Cirrascale customers running large-scale inference, frontier-model training, and fine-tuning, and it extends Cirrascale's open, multi-vendor approach to accelerated computing.

AMD Helios is a rack-scale AI infrastructure platform for frontier AI development and AI factory deployments. It brings AMD Instinct GPUs, 6th Gen AMD EPYC™ CPUs, AMD Pensando™ networking, and AMD ROCm™ software together in a unified, 72-GPU rack-scale platform built from OCP ORW-aligned 4-GPU compute trays. Defense-in-depth security, virtualization, and rack-scale software acceleration round out a system designed on open standards, so customers can deploy and scale interoperable AI infrastructure from a single rack to cluster scale.

At rack scale, AMDHelios delivers exaflop-class AI performance, with up to 2.9 exaFLOPS of peak4-bit (OCP MXFP4) compute and 31 TB of HBM4 memory in a single rack. It scales up inside the rack over UALink™ over Ethernet (UALoE) and scales out across racks and clusters on standards-based Ethernet aligned with the Ultra Ethernet Consortium (UEC), giving customers a path to grow without proprietary fabric lock-in.

The broader AMD Instinct MI400 Series also includes AMD Instinct™ MI430X GPUs, built for sovereign AI and HPC. Combining leadership performance for simulation and modeling with advanced AI capabilities in open, trusted systems, MI430X GPUs give governments, research institutions, and national labs a path to scale private AI compute with control and confidence.

Read the full press release [here](https://www.cirrascale.com/press/cirrascale-amd-instinct-mi400-series-gpus).

[Cirrascale](https://www.cirrascale.com/)

[Sponsored![How Managed Services Help to Reduce Technical Debt](./assets/img-1168dd5c.webp)\\

Jan 26, 2026](https://www.lightreading.com/services/how-managed-services-help-to-reduce-technical-debt?recipe=sponsored-items&source_content_id=4130b2cbc560ce258a2732e20f2b445f) [AI & Machine Learning![2025 in review: Headcount on the slide](./assets/img-553aff43.webp)\\

Dec 22, 2025](https://www.lightreading.com/ai-machine-learning/2025-in-review-headcount-on-the-slide?recipe=similar-items&source_content_id=10395a5dc79bb10fea1c956055715f1c) [AI & Machine Learning![Eurobites: Nokia warns networks need to evolve in AI era](./assets/img-a9ab09ab.webp)\\

Dec 16, 2025](https://www.lightreading.com/ai-machine-learning/eurobites-nokia-warns-networks-need-to-evolve-in-ai-era?recipe=similar-items&source_content_id=929394fb0549bb9a65d7967d4bfa6607) [AI & Machine Learning![Fiber broadband customers generate more frequent, intense AI use – survey](./assets/img-c4e28f47.webp)\\