答案已發布上週Last edited 上週22 來源

Databricks 同 NVIDIA 加深合作：Vera CPU、無伺服器 GPU，仲有成個 Agentic AI 架構

Databricks 同 NVIDIA 合作推出專為「智能代理時代」而設嘅端到端平台，整合咗 NVIDIA 最新 Vera CPU、Blackwell/Rubin GPU、無伺服器 GPU 運行環境同代理專用工具。 Vera CPU 係全球首粒專為 AI 代理而設嘅 Arm 架構處理器，內置 256 MB L3 快取同硬體 JSON 解析單元，專門處理代理工作流程嘅規劃、工具呼叫同記憶循環。

使用 Studio Global AI 搜尋並查核事實瀏覽更多熱門頁面

168K0

Abstract representation of Databricks and NVIDIA partnership for AI agent workloads, featuring data pipelines and GPU acceleration. — How are Databricks and NVIDIA deepening their partnership to optimize AI agent workloads, and what specific technologies, integrations, andDatabricks and NVIDIA are integrating Vera CPUs, serverless GPUs, and agent toolkits into a unified platform.
AI 提示
Create a landscape editorial hero image for this Studio Global article: How are Databricks and NVIDIA deepening their partnership to optimize AI agent workloads, and what specific technologies, integrations, and. Article summary: Databricks and NVIDIA are significantly deepening their partnership to build an end-to-end platform purpose-built for the "agentic era," integrating NVIDIA's latest silicon (Vera CPU, Blackwell/Rubin GPUs), serverless GP. Topic tags: general, general web, user generated, documentation. Style: premium digital editorial illustration, source-backed research mood, clean composition, high detail, modern web publication hero. Use reference image context only for broad subject, composition, and topical grounding; do not copy the exact image. Avoid: logos, brand marks, copyrighted characters, real person likenesses, fake screenshots, UI text, readable text, watermarks,
openai.com

Databricks 同 NVIDIA 最近宣布「加碼」合作，目標係建構一個由底到頂、專為「智能代理時代」（Agentic Era）而設嘅統一平台。呢個平台會直接將 NVIDIA 最新嘅晶片（包括 Vera CPU、Blackwell 同 Rubin GPU）、無伺服器 GPU 運行環境，以及代理專用軟件工具，整合到 Databricks 嘅 Data Intelligence Platform 入面。

成個合作涵蓋 AI 代理嘅成個生命週期——由用無伺服器 NVIDIA GPU 做訓練同微調，到用 NVIDIA Vera CPU 做推理同編排，全部都會黐住 Databricks 嘅 lakehouse 數據治理方案。以下係具體嘅技術、整合方向同業界趨勢。

重點技術同產品整合

NVIDIA Vera CPU：專為 AI 代理而生嘅處理器

Databricks 會將 NVIDIA 新推出嘅 Vera CPU——被稱為「AI 代理嘅 CPU」——整合到平台入面。Vera 係一款自研嘅 Arm 架構 CPU，內置 256 MB L3 快取（比市面上最大嘅 x86 伺服器 CPU 大約 4 倍），仲有一個專用硬體單元處理 JSON 解析，目的係專攻 AI 代理工作流程入面最「塞車」嘅編排同強化學習循環。呢個動作意味住 NVIDIA 同 Databricks 睇好一個新趨勢：AI 代理嘅效能瓶頸唔再係 GPU 本身，而係負責規劃、工具呼叫同記憶循環嘅主機 CPU 。

Studio Global AI

Search, cite, and publish your own answer

Use this topic as a starting point for a fresh source-backed answer, then compare citations before you share it.

使用 Studio Global AI 搜尋並查核事實

人們還問