d-Matrix Emerges From Stealth With Strong AI Performance And Efficiency

November 19, 2024

Karl Freund, Founder and Principal Analyst, Cambrian-AI Research LLC:

Startup launches “Corsair” AI platform with Digital In-Memory Computing, using on-chip SRAM memory that can produce 30,000 tokens/second at 2 ms/token latency for Llama3 70B in a single rack.

d-Matrix uses a hybrid approach to memory that appears to deliver excellent results, using SRAM as “Performance Memory” and a larger DRAM store for “Capacity Memory”. Use the Performance Memory for on-line operations that require low-latency for interactivity, and use the Capacity Memory for off-line work.

Read full article on Forbes

Suggested Articles

Deep divers off pier

Impact of the DeepSeek Moment on Inference Compute 

By d-Matrix Team | January 31, 2025

The Complete Recipe to Unlock AI Reasoning at Enterprise Scale

By d-Matrix Team | February 13, 2025

Think more vs. Train more

By Sid Sheth | January 29, 2025