Hardware & Compute
Chips, data centres and the infrastructure behind AI.
-
The SRAM cost trap in wafer-scale training
Prioritizing on-chip SRAM for massive parameter counts can cause the Cerebras CS-3 TCO advantage to vanish. Storing a Qwen3-235B model with 131K…
-
Tenstorrent Blackhole architecture and AI inference economics
Tenstorrent’s Blackhole architecture offers scalable AI inference for edge deployments, providing a cost-effective alternative to Nvidia hardware. Strategic investments from LG and…
-
The history of NVIDIA’s DGX-1 and the evolution of AI clusters
NVIDIA’s 2016 DGX-1 system used Pascal architecture to deliver the performance of 250 x86 servers in a single 3U box. This breakthrough…
-
The economics of CoreWeave’s capital expansion and GPU pricing
CoreWeave secured over $30 billion in capital to fund rapid expansion, facing rising operating expenses of $1.19 billion. Specialized GPU rental rates…
-
Geopolitical drivers of sovereign compute and national AI programs
Global sovereign AI spending is projected to exceed $100 billion in 2026 as nations build state-owned supercomputers. Major initiatives like Saudi Arabia’s…
-
SambaNova reaches $11 billion valuation driven by inference demand
SambaNova Systems achieved an $11 billion post-money valuation following a $1 billion Series F round. The company is pivoting toward premium inference…
-
The chip architects behind Groq’s TSP architecture and NVIDIA’s challenge
Groq’s LPU architecture provides 80 TB/s of internal memory bandwidth to challenge NVIDIA’s dominance in AI inference. Founded by Jonathan Ross, the…
-
Cerebras and G42 expand AI supercomputing with Condor Galaxy 3
Cerebras and G42 have launched Condor Galaxy 3, a supercomputing cluster delivering 8 exaFLOPs of AI performance using 64 Cerebras CS-3 systems.…
-
Deploying FuriosaAI RNGD for Korean Cloud Inference
FuriosaAI’s RNGD NPU delivers 2.25x better performance per watt for LLMs than GPU solutions. Samsung SDS hosts these accelerators at its Dongtan…
-
Positron Asimov vs Lightmatter Passage for data center inference
Positron’s Asimov silicon targets 5x more tokens per watt than Nvidia’s Rubin, while Lightmatter’s Passage platform provides up to 256 Tbps of…
-
How to get started with Axelera’s $500M Metis AI edge processor
Axelera’s $500M Metis AI edge processor enables advanced retail loss prevention and factory defect detection pipelines. The technology addresses complex data challenges…
-
Intel diverts resources to Jaguar Shores as AMD gains ground
Intel is pivoting from its Falcon Shores AI accelerator to focus on the Jaguar Shores rack-scale system following a $1.6 billion net…