IBM's 12nm Chip Just Made Nvidia's 4nm GPUs Look Like a JOKE!
By Evolving AI
Community Score: 50% | 412.6K views | 4w
0 community ratings: null thumbs up, null thumbs down
IBM’s NorthPole AI chip could point toward a completely different future for artificial intelligence. Instead of relying on bigger GPUs, faster HBM, and enormous amounts of power, IBM designed NorthPole to bring memory and AI compute together on the same chip, attacking the decades-old von Neumann bottleneck responsible for much of the data movement in conventional computing. The IBM NorthPole chip is built on a 12nm process with 22 billion transistors, 256 cores, and roughly 13 TB/s of on-chip memory bandwidth. In IBM’s image-recognition comparisons, NorthPole delivered around 22× more work per joule than the GPUs it was tested against. IBM also demonstrated NorthPole running Granite AI models, including a 16-chip system exceeding 28,000 tokens per second while using about 672 watts of card power. In this video, we break down how IBM NorthPole works, its brain-inspired memory architecture, on-chip SRAM, AI inference performance, and energy efficiency. We also compare NorthPole with
Tags: IBM NorthPole, AI chip, AI chips, AI processor, AI accelerator, AI inference, AI hardware, artificial intelligence, IBM AI, IBM Granite, Granite AI, Nvidia, Nvidia GPU, energy efficient chip, neuromorphic computing, neuromorphic chip, brain inspired chip, AI data center, AI computing, data center AI
More from Evolving AI
- Humanoid Robots Now Cost €2,999 - And They Already Have Jobs! — Score: 50%
- OpenAI Just Got Hacked By an AI in 72 Hours! — Score: 50%
- An AI Trained Itself for Weeks & Beat 3,992 Human Teams - SELF-IMPROVEMENT REVOLUTION! — Score: 50%
- OpenAI Just Revaled Something More Terriying Than AGI - RSI is Here! — Score: 50%
- OpenAI's AI Designed Its Own Chip in 9 Months — And It Beat Nvidia! — Score: 50%
- This New Chip Factory Made NVIDIA & TSMC Look Like a Billions-of-Dollars JOKE! — Score: 50%