inference
-
techIntel Targets Nvidia with Budget AI Chip Crescent Island
Intel plans a new AI chip called Crescent Island by end of 2026, using cheaper memory to undercut Nvidia in inference computing.
-
techNvidia Doubles Down on Agentic AI Inference
As AI shifts from training to deployment, Nvidia’s Blackwell Ultra targets agentic workloads with 50x performance gains.
-
techCerebras Boosts IPO Target to $4.8 Billion Amid Surging Demand
AI chipmaker Cerebras raises IPO price target to $150-160 per share, aiming for $4.8 billion amid strong investor demand.
-
techGoogle Explores Marvell Partnership for Advanced AI Chips
Google is reportedly in discussions with Marvell Technology to design new AI chips focused on enhancing inference capabilities. The move signals a strategic push into specialized AI hardware.
-
techDigital Realty Boosts Singapore AI Infrastructure with S$7 Billion Investment
Digital Realty significantly expands its data center footprint in Singapore, investing up to S$7 billion to support the next phase of AI inference and cloud connectivity.
-
techNvidia Secures Massive AWS AI Chip Deal
AWS to deploy 1 million Nvidia GPUs by 2027, deepening AI infrastructure collaboration amid rising inference demand and global supply scrutiny.