HARDWARE

–

27 October 2023 0h17

This is not a real display

It’s an simulated 2D screen with dynamic lighting inside mixed reality.

🤯

pic.twitter.com/UlODEMSJkW
— Linus ✦ Ekenstam (@LinusEkenstam) 26 octobre 2023

This is not a real display It’s an simulated 2D screen with dynamic lighting inside mixed reality.

27 October 2023

Snapdragon and Apple M2/M3 Running LLMs On-Device

By

–

26 October 2023 23h12

Yes, def, did you see the snapdragon keynote? 30 tokens a second for Llama2 7B param on device. Maximum 13B param available to run on device. Apple can def run LLM’s on M2/M3

Future Head-Mounted Devices: Brutalist Soviet Design Exploration

By

–

26 October 2023 22h52

Prompt Share | Midjourney Digital head mounted devices of the future. Love exploring these odd buckets of wonders. What will you make? A device that goes on your head, 1970's brutalist soviet industrial design, product design photograph –style raw –v 5.2

→ View original post on X — @cerebras

Cerebras Co-Founder Presents AI Hardware Innovations at OktoberTech

By

@cerebras

–

26 October 2023 21h19

Cerebras Co-Founder JP Fricker had a great time presenting at OktoberTech 2023! This annual event by @Infineon Technologies brought together some of the brightest minds in the tech industry to share their perspectives on the latest trends and innovations. #Digitalization

→ View original post on X — @nvidiaai

Google Cloud G2 VMs NVIDIA L4 GPU AI Model Deployment

By

@nvidiaai

–

26 October 2023 18h00

In this @googlecloud webinar, explore how G2 VMs powered by NVIDIA L4 Tensor Core GPUs and Triton Inference Server can standardize #AI model deployments across your workloads and how @Get_Writer is using G2 VMs to bring #generativeAI to enterprises. https://
nvda.ws/473cGvQ

Peak On-Device Model Capacity Reaches 13B Parameters

By

–

26 October 2023 15h00

Currently peak on device is 13B parameters

Llama2-7B Reaches 30 Tokens Per Second On-Device

By

–

26 October 2023 14h59

Here we go. Llama2-7B at 30 tokens per second ON CHIP. Give it 2 more cycles and we’ll see the larger models on device at these speeds. LLMs will be as omnipresent as calculators, simple maths.

→ View original post on X — @mathworks

Altgreen’s Eco-Friendly Electric Three-Wheeler Solution for India

By

@mathworks

–

26 October 2023 12h00

@AltigreenL introduces a more eco-friendly and cost-effective solution for deliveries in India. This electric three-wheeler uses only 12 units of electricity per charge #ModelBasedDesign #ElectricVehicles #Simuliink