AI HARDWARE - AI Dynamics

HiPEAC23: Maxeler accelerates systems with real-time AI solutions

By

–

16 January 2023 21h34

#HiPEAC23 is the premier European forum for computer architecture, programming models, compilers & operating systems for general-purpose, embedded and cyber-physical systems. Stop by our booth to hear how @MaxelerTech can accelerate your systems with real time #AI solutions

→ View original post on X — @groqinc

16 January 2023

AI Hardware Demand Explosion and Tenstorrent’s Progress

By

@akshay_pachaar

–

15 January 2023 4h45

AI hardware demand is bound to explode!!
Tenstorrent is doing an amazing job

→ View original post on X — @akshay_pachaar

15 January 2023

10 Breakthrough Technologies 2023

By

@mfordfuture

–

13 January 2023 23h19

10 Breakthrough
Technologies 2023

→ View original post on X — @mfordfuture

13 January 2023

Energy-Centric AI: Efficient ML Models Without Precision Loss

By

@untetherai

–

13 January 2023 22h40

How can #ML/#DL models run efficiently without precision? In the latest @SemiEngineering article, @UntetherAI
's @beach66 & Martin Snelgrove introduce the Energy-Centric AI computing architecture: a solution built for energy-efficient #AI inference.

→ View original post on X — @untetherai

13 January 2023

Cerebras Wafer-Scale Engine: 850K cores, 40GB SRAM distributed

By

@cerebras

–

13 January 2023 22h01

Some #wafer-scale eye candy When performance is everything, the best #memory hierarchy is no memory hierarchy at all! The Cerebras Wafer-Scale Engine has 40GB of #sram evenly distributed across 850,000 cores for a staggering total of 2.6 trillion #transistors. Thanks, #tsmc!

→ View original post on X — @cerebras

13 January 2023

Organic Sensing Stack and Post Processing Deep Dive

By

@bilawalsidhu

–

13 January 2023 17h16

ROFL you’re getting deep with our own organic sensing stack and post processing therein

→ View original post on X — @bilawalsidhu

13 January 2023

GroqChip runs 4x faster than Nvidia V100 on CFD

By

@groqinc

–

12 January 2023 18h37

See how GroqChip™ runs 4x faster than Nvidia V100 GPU on a computational fluid dynamics application in this @MaxelerTech demo from #SC22. Learn how converged HPC may offer a faster time-to-solution and compute efficiency without sacrificing accuracy. https://
youtu.be/KDP7tx9QYT0

→ View original post on X — @groqinc

12 January 2023

Anticipation for Large Model with Cost and Speed Concerns

By

@thatroblennon

–

12 January 2023 17h55

Yep, it's gonna be huge. Hope the model isn't so big that the processing isn't too cost-prohibitive or slow

→ View original post on X — @thatroblennon

12 January 2023

MineDojo: Training AI Agents in Minecraft Open-Ended Tasks

By

@nvidia

–

12 January 2023 17h45

To boost AI research, we helped create MineDojo. It uses the almost infinite number of open-ended tasks and actions in Minecraft to train #AI agents. And for the gamers who love that world, @Minecraft with RTX lights it up like never before. https://t.co/sf8g0wmnBK #NVIDIAstory
— NVIDIA (@nvidia) 12 janvier 2023

To boost AI research, we helped create MineDojo. It uses the almost infinite number of open-ended tasks and actions in Minecraft to train #AI agents. And for the gamers who love that world, @Minecraft with RTX lights it up like never before. https://
nvda.ws/3xTkalW #NVIDIAstory

→ View original post on X — @nvidia

12 January 2023

Generative AI Investment Surge Drives Data Center Hardware Innovation

By

@cerebras

–

11 January 2023 23h25

Data Center Frontier mentions that "Surging interest in generative AI will lead to fresh investment…[creating] opportunities for… offerings optimized around new processors (like CirraScale’s collaboration with Cerebras)." Article:

→ View original post on X — @cerebras

11 January 2023