AIINEWS

AIINEWS

Global intelligence on artificial intelligence and computing

D-Matrix builds Raptor chip with stacked 3D DRAM to tackle AI inference memory limits

AIINEWS Editorial Board•hardware
20 November, 2026•2 min read
D-Matrix builds Raptor chip with stacked 3D DRAM to tackle AI inference memory limits

Executive Briefing

“D-Matrix is developing a next-generation chip called Raptor that stacks 3D DRAM beneath its...”

Key Takeaways

  • 1.D-Matrix is answering with Raptor, a second-generation chip stacking 3D DRAM beneath its compute logic to address memory bandwidth during artificial intelligence inference.
  • 2.According to d-Matrix fellow Satyam Srivastava, the stacked 3D DRAM architecture places the technology in a new tier sitting between SRAM and high-bandwidth memory.
  • 3.D-Matrix has also adopted NVLink Fusion rackscale infrastructure from Nvidia Corp. to connect its accelerators to Nvidia’s AI factories.

D-Matrix is developing a next-generation chip called Raptor that stacks 3D DRAM beneath its compute logic to overcome memory bandwidth bottlenecks during artificial intelligence inference workloads.

Product Features & Key Highlights

D-Matrix is answering with Raptor, a second-generation chip stacking 3D DRAM beneath its compute logic to address memory bandwidth during artificial intelligence inference.

According to d-Matrix fellow Satyam Srivastava, the stacked 3D DRAM architecture places the technology in a new tier sitting between SRAM and high-bandwidth memory.

D-Matrix has also adopted NVLink Fusion rackscale infrastructure from Nvidia Corp. to connect its accelerators to Nvidia’s AI factories.

Practical Assessment & Evaluation

Placing 3D DRAM directly beneath the logic layer shortens data travel distances without crossing an interposer, thereby optimizing energy efficiency and bandwidth for hardware systems.

This integration strategy reflects the trend toward heterogeneous computing, where server clusters coordinate flexibly among GPUs, CPUs, and specialized accelerators to maximize overall system performance.

Pricing, Availability & Next Steps

Detailed technical metrics regarding actual power consumption, tokens-per-watt performance, and commercial pricing for the Raptor chip have not been fully disclosed.

The specific commercial release timeline and integration capabilities of the Raptor chip in large-scale data centers beyond initial test assertions require further validation.

Independently verified, edited, and published by AIINEWS Editorial Board
3-tier verified
Editorial standards
Share this intelligence report:
Expanded Perspectives & Intelligence

Related Dispatches & Deep Dives

Curated 8 specialized reports
Clipto enables ChatGPT and Claude to search and analyze years of people’s meetings and video calls

Clipto enables ChatGPT and Claude to search and analyze years of people’s meetings and video calls

Artificial intelligence assistants such as ChatGPT and Claude have become essential tools for millions of workers worldwide, helping them compose emails in seconds, generate complex code and draft reports by looking at whatever data that person has access to. But there’s one thing that’s missing from those chatbots’ knowledge. Ask them what was discussed in […] The post Clipto enables ChatGPT and Claude to search and analyze years of people’s meetings and video calls appeared first on SiliconANGLE.

5 min read7,958

Explore AIINEWS Intelligence Vault

Over 2,000 in-depth dispatches structured across 18 specialized tech & AI categories.

Browse all 18 categories