News World That's Worth

Google Partners with Blackstone in Landmark Joint Venture to Offer Dedicated Wholesale TPU Leases In a massive structural shift designed to reshape the high-performance AI computing market, Google has announced a joint venture with Blackstone , the world’s largest alternative asset manager with over $1.3 trillion in assets under management. The primary objective of this new enterprise is to provide global enterprise clients with an alternative route to lease Google’s proprietary Tensor Processing Units (TPUs) . This move bypasses the traditional cloud-credit operational model of Google Cloud, offering a wholesale, flat-rate hardware leasing framework specifically optimized for hyper-scale corporations and institutional AI developers requiring immense, uninterrupted compute blocks. The Financial and Operational Blueprint While the new joint venture has yet to receive an official corporate name, Blackstone has committed an initial seed investment of $5 billion to kickstart the project...

Google Unveils TPU v8: A Dual-Chip Architecture for the Future of AI Google has officially unveiled its 8th-generation Tensor Processing Unit (TPU) architecture, marked by a strategic split into two distinct, specialized chips: the TPU v8t (for training) and the TPU v8i (for inference). While they share a brand name, they are built on fundamentally different architectures designed to solve the distinct bottlenecks of model training and large-scale deployment. TPU v8t: The Training Powerhouse Engineered for massive-scale training, the v8t is a monster of performance. SparseCore & FP4: It introduces a dedicated SparseCore for memory access control and an MXU (Matrix Multiplication Unit) with native FP4 (4-bit floating point) support, doubling down on efficiency. Virgo Network: It features the new Virgo Network , capable of interconnecting up to 134,000 chips at a staggering total bandwidth of 47 Petabits per second , delivering a peak performance of 1.6 ExaFLOPS . Direct Pat...

Anthropic Scales for the Future: Multi-Gigawatt Google Cloud TPU Deal to Power Next-Gen Claude Models Anthropic has announced a major expansion of its partnership with Google Cloud , securing massive computational capacity to power the development of its next-generation Claude AI models. The agreement centers on the use of Google Tensor Processing Units (TPUs) , scaling up to a combined power capacity of several gigawatts . This infrastructure boost is designed to meet surging global demand and support the highly anticipated hardware expected to go live by 2027. The Broadcom Connection A key player in this infrastructure play is Broadcom , the primary chip manufacturer for Google’s custom TPUs. Broadcom confirmed its role in this expansion, noting that the total processing power it currently supports for its custom silicon clients has reached approximately 3.5 gigawatts . Explosive Revenue and Enterprise Adoption The scale of this expansion is driven by Anthropic's staggering fin...