#ai infrastructure
Ai Infrastructure: 131 AI articles covering ai infrastructure news, analysis, and research
Articles
What Is Decision 3.0? Inside vLLM Semantic Router's New Openβ7
vLLM Semantic Router releases Decision 3.0, a family of five Apache-2.0 multimodal decision models from 0.8B to 27B that classify, route, and gate requests.
Can the AI Industry Win Over Data Center Opponents by Ditching NDAs?β9
Amazon and Microsoft are dropping NDAs in data center negotiations, but will it quell growing local opposition? We discuss on the Equity podcast.
Nace AI Open-Sources Drex 1.5: A 9B Decision Model That Scoresβ8
Nace AI open-sourced Drex 1.5, a 9B decision model for agents that scores options instead of generating text. It tops the Decision Index under 10B parameters.
Amazon Drops Data Center NDAs, While AI Agents Push to Accessβ9
Amazon drops data center NDAs as community backlash grows, while AI startups race to get consumers to trust agents with credit cards, inboxes, and files.
Amazon and Others End Secret Data Center Deals: Is It Enough toβ9
Amazon and Microsoft end NDAs on data center deals amid community backlash, as AI startups bet consumers will trust agents with inboxes, files, and credit cards...
Impactful Scheduling for GPU Clusters: Efficient Orchestration in 2026β9
Discover key strategies for impactful GPU cluster scheduling in 2026, including priority-aware, topology-aware, and dynamic resource pooling for AI workloads.
Ambrosia Energy and Bloom Energy Execs to Headline AIβ8
Ambrosia Energy and Bloom Energy execs discuss AI's physical infrastructure bottlenecks, energy solutions, and where founders can build at TechCrunch Disrupt 20...
Architect Launches Liquid Inference: A Real-Time Auctionβ9
Architect launches Liquid Inference, a real-time auction marketplace where LLM providers bid per request, letting developers cut costs by swapping a base URL wi...
Perplexity AI Unveils pplx-embed-v2-late: A 0.6B Edge Model andβ7
Perplexity releases pplx-embed-v2-late: 0.6B and 9B ColBERT-style multimodal embedding models for text, image, and PDF retrieval, scoring 92.4% on MADQA.
Meta AI Open-Sources Rebalancer: A C++ Assignment Solverβ10
Meta AI open-sources Rebalancer, a C++ assignment solver handling 40 million daily placement problems across large-scale AI infrastructure.
