d-Matrix turns its inference chips into 1,024-XPU data-center clusters

D-Matrix announced Squadron compute clusters that scale its Corsair inference accelerators to 1,024 XPUs per data-center deployment, with founder and CEO Sid Sheth saying the clusters are shipping to priority customers. The company-reported demo paired Corsair accelerators with GPUs, and d-Matrix is selling a complete inference cluster rather than an accelerator alone, extending Sheth's 2019 bet into data-center infrastructure. Customer deployment and independently comparable performance remain the tests.

d-Matrix turns its inference chips into 1,024-XPU data-center clusters d-Matrix founder Sid Sheth says Squadron is shipping to priority customers; a company-reported demo paired Corsair accelerators with GPUs. By RuntimeWire Staff https://runtimewire.com/author/runtimewire-staff · Published Primary source: PR Newswire https://www.prnewswire.com/news-releases/d-matrix-announces-squadron-compute-clusters-for-ai-inference-at-data-center-scale-302906041.html Why it matters d-Matrix is selling a complete inference cluster rather than an accelerator alone, extending Sid Sheth's 2019 bet into data-center infrastructure. Customer deployment and independently comparable performance remain the tests. Sid Sheth https://www.d-matrix.ai/transforming-ai-d-matrixs-pivotal-moments-in-pursuit-of-gen-ai-inference-at-scale/?ref=runtimewire , the co-founder and CEO of d-Matrix https://d-matrix.ai/?ref=runtimewire , spent years arguing that inference would become a larger compute opportunity than training.…