Pinned post

As inference splits into prefill and decode, Nvidia's Groq deal could enable a "Rubin SRAM" variant optimized for ultra-low latency agentic reasoning workloads (Gavin Baker/@gavinsbaker)

Gavin Baker / @gavinsbaker : As inference splits into prefill and decode, Nvidia's Groq deal could enable a “Rubin SRAM” variant optim...

11 December 2025

Rivian says it is building its own 5nm AI chip called the Rivian Autonomy Processor and a foundational "Large Driving Model", to enable fully autonomous driving (Andrew J. Hawkins/The Verge)

Andrew J. Hawkins / The Verge:
Rivian says it is building its own 5nm AI chip called the Rivian Autonomy Processor and a foundational “Large Driving Model”, to enable fully autonomous driving  —  The company's chips, quoted at 1,600 trillion operations a second, will fuel a new push into self-driving cars for the EV company.

Posted from: this blog via Microsoft Power Automate.

Daily Deals

By on