Former Lambda, Groq alums secure $875M to build ‘memory-first’ AI inference chips
Neocloud veterans bank on low-power DRAM to power custom inference servers
News - AI inference
The latest news on networking, cloud, and AI infrastructure
-
AMD, Cerebras partner on joint Helios rack-scale AI inference platform
Literal chip giant to integrate Wafer-Scale Engine in Helios racks for enhanced AI inferencing
-
AMD acquires FastFlowLM to accelerate on-device AI inferencing
Open alternative to CUDA bolstered by AI optimization for client and workstation
-
Microsoft ramps hiring for photonic AI interconnect push
Hyperscale wants photonics boffins to wire up its AI infrastructure with light
-
Hyperscalers lead Gartner’s AI infra ranking; neoclouds step up as challengers
First-ever cloud AI infrastructure report pits hyperscale giants against agile, purpose-built neocloud alternatives
-
Out now: SDxCentral's Storage Supplement
Storage has never been sexier
-
Cornelis Networks, NextSilicon partner to accelerate AI and HPC infrastructure
Combating infrastructure bottleneck aim of effort while scale-out networking solution validated in Stampede3 supercomputer
-
Nvidia and AMD plow $20M in AI inference disruptor
Tensormesh now worth $24.5 million and counting
-
China Mobile latest operator to sell AI tokens to Chinese consumers
For under $6, customers can now top up their phones to power agentic workloads
-
Nvidia gets greener, edging 200% networking revenue bump
Jensen claims supply chain remains sturdy while predicting an AI-RAN future
-
AI cloud DeepInfra raises $107M in Series B funding round
Will be used to scale its cloud platform
-
Zero Latency taps Red Hat and Nvidia for US-wide distributed AI infrastructure
Red Hat’s AI Factory platform to act as enterprise Kubernetes foundation to route workloads across edge infrastructure
-
NTT in all-photonics pivot as earnings look beyond mobile
DoCoMo performance sees Japanese giant push back targets and go all in on IOWN APN
-
Zero Latency launches beta for AI inference orchestration platform
Zerogrid routes inference workloads according to latency, locality, and capacity constraints
-
Viavi test platform brings security to inference and AI data centers
CyberFlood CF1000 Appliance also touts quantum secure credentials
-
Google Cloud cozies up to Nvidia’s Vera Rubin for new bare metal A5X AI inference VMs
Google’s Virgo fabric to connect A5X instances, scaling AI up to 960,000 distributed GPUs
-
DDN, Google Cloud claim Lustre KV cache trick boosts AI inference throughput 75%
Managed Lustre gets inference upgrade as customers move from training to deployment
-
Adtran amps up AI on the edge with 400G-ready routers
'Terabit-class' releases to solve rising mismatch between real-world requirements and oversized platforms
-
Samsung serves frontier cloud AI with leading inference player
FriendliAI to power SCP for South Korea and beyond
-
Gartner: LLMs to be up to 100X more cost-efficient by 2030
But enterprises are warned not to skimp with cheap tokens
-
Nvidia, hyperscaler-backed open standard for AI inference torch passed to Linux Foundation
Supported by the likes of Google Cloud and CoreWeave, LLM-D gets sandbox status from CNCF
-
Introducing the SDxCentral Magazine
Two firsts at once: SDxCentral launches debut magazine at the Xcelerated Compute Show
-
SK Telecom to solve AI memory blight with Hynix in 2026
APAC giant gives rare insight outside of South Korea into its operations since recent AI-focused restructure
-
AWS partners with big chip co. Cerebras for AI 'inference disaggregation'
Will be deployed on Amazon Bedrock in AWS data centers
-
Equinix launches 'vendor-neutral' Distributed AI Hub to bring inference closer to the edge
Platform offers a single, unified framework for connecting to GPU clouds and data platforms