home hero bg

Low latency AI inference without compromises

d-Matrix deploys revolutionary technology in memory-centric compute, next-generation I/O, and stacked DRAM solutions to power low latency AI inference at scale.

d-Matrix and Gimlet Labs to Deliver 10x Inference Speed Up and Power Efficiency

Corsair + GPU disaggregated pipelines deliver 10x the performance of a standard GPU-only pipeline.

A radically different approach to compute + memory

  • Memory-centric approach prevents latency bottlenecks to deliver low-latency interactive applications.
  • Chiplet-based design enables scaling SRAM-based architecture to power models up to 100B parameters.
  • PCIe form factor delivers instant results with existing data center configurations.

Who is d-Matrix?

d-Matrix
We’re industry veterans, who’ve shipped over 100 million chips. Years before generative AI started captivating imaginations, we were already at work— quietly making bold moves to take AI farther than anyone else.

Inspired by the visionaries, the ones who think different, who are dissatisfied with the status quo, who dare to dream a different future and then go ahead and build it.

Sure, we seemed like round pegs in a square hole, but that’s because nobody was able to see what we did.


Meet Corsair™, the world’s most efficient AI inference computing platform for datacenters

Scale with JetStream™, purpose-built I/O accelerator for AI inference

A next-generation blazing fast accelerator-to-accelerator communications platform that scales up to millions of requests.

Announcing SquadRack™, industry’s first rack-scale solution, purpose-built for AI inference, using a disaggregated standards-based approach. Built with industry-leading AI infrastructure providers.

Redefining Performance and Efficiency for AI Inference at Scale

Blazing Fast

x

interactive-speed

Commercially Viable

x

cost-performance

Sustainable

x

energy-efficiency

Performance projections for Llama-70B, 4K context length, 8-bit inference vs H100 GPU, results may vary

Built without compromise

Don’t limit what AI can truly achieve and who can benefit from it. We’ve built Corsair from the ground up, with a first-principle approach . Delivering Gen AI without compromising on speed, efficiency, sustainability or usability.

Performant AI

d-Matrix delivers ultra-low latency high batched throughput, making Enterprise workloads for GenAI efficient

Sustainable AI

AI is on an unsustainable trajectory with increasing energy consumption and compute costs. d-Matrix let’s you to do more with less.

Scalable AI

Our purpose-built solution scales across models and is enabling enterprises and datacenters of all sizes to adopt GenAI quickly and easily.

Performant AI

d-Matrix delivers ultra low latency without compromising throughput, unlocking the next wave of Generative AI use cases

fighter jet sound cloud

Sustainable AI

AI is on an unsustainable trajectory with increasing energy consumption and compute costs. d-Matrix enables you to do more with less.

sustainable ai spot

Scalable AI

d-Matrix offers a purpose-built solution that scales with model size to empower companies of all sizes and budgets.

scalable ai optimized

Blazing fast

Commercially viable

Energy efficient