AWS launches llm-d disaggregated inference on SageMaker HyperPod and EKS, delivering up to 70% higher LLM throughput by splitting compute-heavy prefill and memory-bound decode phases across distributed GPUs using an open-source Kubernetes-native framework built on vLLM.
Meta commits up to $27 billion to cloud provider Nebius Group over five years, securing massive AI infrastructure capacity as it races to compete in the frontier AI model race.
Netflix engineers discover a Linux kernel lock bug causing container stalls lasting tens of seconds, tracing the issue to global mount lock contention triggered by thousands of concurrent bind mount operations, with hardware architecture playing a critical role and mitigations already underway.
Claude's Excel and PowerPoint add-ins now share full context across all open files in one continuous conversation, offer one-click Skills for tasks like LBO models and pitch deck reviews, and are deployable across Amazon Bedrock, Google Vertex AI, and Microsoft Foundry.
Cloudflare launches a powerful new /crawl endpoint in open beta, enabling developers to crawl entire websites with a single API call and retrieve content in HTML, Markdown, or structured JSON formats, with advanced controls like depth limits, URL filtering, and incremental crawling available on both free and paid plans.
Google is rolling out sweeping Gemini AI upgrades across Workspace, transforming Docs, Sheets, Slides, and Drive into intelligent productivity tools that generate drafts, populate data 9x faster than manual entry, and turn Drive into an active AI-powered knowledge base pulling answers from files, Gmail, Calendar, and Chat.
Iran strikes AWS data centers in the UAE and Bahrain, disrupting banking and consumer services while casting a shadow over billions in regional AI investments, including Microsoft's $15 billion UAE commitment, as companies now grapple with the reality that data centers have become targets in modern armed conflict.
Google is rolling out powerful Gemini AI upgrades across Docs, Sheets, Slides, and Drive, letting subscribers generate full drafts, build spreadsheets from prompts, and create presentations by pulling content from Gmail, Drive, and the web.
Google launches Gemini Embedding 2, its first natively multimodal embedding model, capable of processing text, images, video, audio, and documents together in a single unified space across 100+ languages, now available in public preview via the Gemini API and Vertex AI.
Apple's M5 Max MacBook Pro, with 128GB unified memory, is enabling developers to run massive 70-billion-parameter AI models locally for free as cloud AI inference costs spiral out of control, with some startups reporting costs tripling in just three months.