d-Matrix Inc.

d-Matrix powers next-generation compute for generative AI inference, having redefined silicon architecture with memory-centric technology that brings compute directly into memory. The company is led by a team of entrepreneurs with a 20+ year history in building businesses that have shipped over 100 million chips and generated over $5 billion in revenue. Its leadership includes Founder & CEO Sid Sheth and Founder & CTO Sudeep Bhoja. d-Matrix has announced a partnership with NVIDIA on NVLink Fusion to connect next-generation d-Matrix XPUs to the NVIDIA AI platform, and it introduced Corsair, a product purpose-built for AI inference that brings high-bandwidth memory and compute together on the same fabric with an inference-optimized compute engine, high-speed chip-to-chip connectivity, and an industry-standard PCIe form factor.

d-Matrix describes itself as industry veterans who have shipped over 100 million chips, and states that years before generative AI started captivating imaginations it was already at work on AI. The company set its sights on building a faster, attainable, more energy efficient AI compute platform that everyone can use, pushing beyond current limitations to introduce a new era of AI. It has redefined silicon architecture with memory-centric technology, moving compute into memory to collapse the distance between memory and compute and accelerate AI inference. d-Matrix announced a partnership with NVIDIA on NVLink Fusion to connect next-generation d-Matrix XPUs to the NVIDIA AI platform, and introduced Corsair, purpose-built for AI inference.

Memory-centric compute for AI inference, built to scale without compromise.

• Corsair

  Purpose-built for AI inference, Corsair brings high-bandwidth memory and compute together on the same fabric, with an inference-optimized compute engine, high-speed chip-to-chip connectivity, and an industry-standard PCIe form factor.

• Unconfirmed

  NVIDIA plans to invest in AI server chip maker d-Matrix[3]英伟达拟投资AI服务器芯片企业d-Matrixfinance.eastmoney.com

  According to Reuters citing three people familiar with the matter, NVIDIA plans to invest in d-Matrix, a seven-year-old company that makes AI server chips and is itself a NVIDIA competitor. The proposed investment is part of NVIDIA's effort to make its technology compatible with competing chip designers and bring them into its hardware ecosystem, and is NVIDIA's first investment in an AI inference chip startup since its US$20 billion licensing agreement with Groq last December.

Read more

1. 1d-Matrix - Memory-Centric Compute for AI Inferenced-matrix.ai
2. 2About d-Matrix | Innovators in AI Compute Solutions & Technologyd-matrix.ai
3. 3英伟达拟投资AI服务器芯片企业d-Matrixfinance.eastmoney.com
4. 410月10日美股成交额前20：微软涨2.38% 推出面向快速决策的Microsoft-Decision-1模型finance.sina.com.cn