Your Agent's Cache Hit Rate Is Lying to You
Independent audits of real production traffic put physical KV cache hit rates in the 70s on average, while the logical hit rates reported in agent traces can run 90% to 99%, a gap of twenty to thirty …
Independent audits of real production traffic put physical KV cache hit rates in the 70s on average, while the logical hit rates reported in agent traces can run 90% to 99%, a gap of twenty to thirty …
Delos Data announced at the AI Infra Summit in Santa Clara, California, that it is developing Apollo, a data interface chiplet offered as a 30+ Tbps I/O chiplet next to an XPU, a 10+ Tbps near-package…
Meta built and operates Prometheus, a one-gigawatt-scale AI training supercluster in New Albany, Ohio, spanning thousands of acres with dozens of buildings and hundreds of thousands of GPUs, and now p…
Seagate displayed 44TB HAMR hard drives and Western Digital outlined its HDD roadmap at the 2026 AI Infra Summit, according to Forbes contributor Thomas Coughlin. The storage vendors presented high-ca…
Companies are granting AI agents access to files, software and corporate data faster than they are building controls to govern what those agents can see, inherit and reach, governance executives said …
At the AI Infra Summit, AMD argued that AI infrastructure should be designed around adaptability rather than any single chip or fabric, citing three auditable properties — adaptable, economical and op…
Data center builders at the AI Infra Summit in Santa Clara on Sept. 16, 2026 said the industry's future hinges less on AI itself than on securing electricity, grid capacity and community trust, accord…
AI Infra Summit in Santa Clara drew 8,000 attendees on day one, up from 3,500 last year, with organizers framing the event around the message that "the age of inference is here, and it's time to conne…
NVIDIA vice president of hyperscale and high-performance computing Ian Buck announced at the AI Infra Summit in Santa Clara that NVIDIA's DSX MaxLPS can deliver up to 1.4x more tokens per megawatt thr…
Lightbits Labs unveiled Inferra, a KV cache orchestration engine it says delivers up to 16x more concurrent inference sessions on existing GPUs, over 100x latency improvement versus recomputing attent…
Cognichip unveiled Artificial Chip Intelligence (ACI), a physics-informed AI system for chip design that founder and CEO Faraj Aalaei says compresses months of front-end work into days, according to a…
The Bay Area events calendar for Sept. 11-17 lists community activities including the AI Infra Summit, a three-day event starting Sept. 15 at the Santa Clara Convention Center, where attendees can lea…
Intel will showcase its AI infrastructure vision at AI Infra Summit 2026, September 15–17 at the Santa Clara Convention Center, with CEO Lip-Bu Tan delivering a fireside chat on September 15. The comp…