Microsoft-backed D-Matrix enters production with Corsair AI inference chip claiming 10x speed over Nvidia GPUs
By
Katie Tarasov
Pure flour-power. Hearty enough to carry you through lunch.
Summary
D-Matrix, a startup based near Nvidia's Silicon Valley headquarters and backed by Microsoft, is entering full production of its Corsair AI inference chip. The company claims Corsair can run inference workloads 10 times faster and use five times less energy than a standalone Nvidia GPU, though only for smaller workloads. The chip takes a novel approach to memory similar to competitors Cerebras and Groq, aiming to bypass the memory shortage that has constrained AI hardware development.
Key quotes
· 3 pulledD-Matrix says its chips can run inference workloads 10 times faster and using five times less energy than a standalone graphics processing unit from the market leader — as long as the workloads are small.
The new inference chip, called Corsair, takes a novel approach to memory that's similar to Cerebras and Groq.
With tech giants demanding all the computing resources
You might also wanna read
NVIDIA DGX Spark Review: Compact Workstation for High-Performance AI Inference
The article provides an in-depth review of NVIDIA's DGX Spark system, an unconventional compact workstation that brings supercomputing-class
General Compute Launches ASIC-Based Inference Cloud for Faster AI Agent Performance
General Compute is an inference cloud built on ASICs (purpose-built alternatives to Nvidia GPUs) designed specifically for AI inference, not
Nvidia Acquires AI Chip Company Groq Despite 75% Revenue Shortfall
Nvidia acquired Groq (with a Q), a company specializing in AI chip technology for faster LLM inference, despite Groq missing its revenue tar
Kog AI Launches Inference Engine Tech Preview: 3,000 Tokens/s on AMD MI300X GPUs
Kog AI launches a tech preview of the Kog Inference Engine (KIE), achieving 3,000 output tokens/s per request on 8× AMD MI300X GPUs and 2,10
blog.kog.ai·12d agoNVIDIA DGX Performance Analysis: Benchmark Results vs Real-World Applications
This article appears to be a technical benchmark comparison between NVIDIA DGX lab performance and real-world applications, likely focusing

Nvidia's DGX Spark Personal AI Supercomputer Launches October 15th at $3,999
Nvidia is launching its DGX Spark 'personal AI supercomputer' on October 15th, a compact desktop machine designed for sophisticated AI model
