COMPUTING

→ View original post on X — @fogoros

Unlocking IIoT Device Potential: Industrial IoT Strategy

By

@fogoros

–

11 May 2024 16h00

Join us at #IIoTWorldDay for "Unlocking Potential: Exploring Your Most Valuable IIoT Devices," moderated by Hamish Mackenzie. With a proven record in tech strategy, Hamish's insights promise valuable moderation. Register now https://
buff.ly/49lF0dW sponsored #cirrus_iiot #ge_iiot

WebGPU capabilities and browser support status check

By

–

11 May 2024 14h48

Not sure but check CANIUSE webGPU

GPUs Already Standard on Most Devices Now

By

–

11 May 2024 13h48

It already exists, most devices have GPUs now

Smaller Models and Bigger GPUs Reshape AI Hardware Future

By

–

11 May 2024 13h44

Yep, but models will get smaller and GPUs bigger esp now that AI is so popular I think

GPU and Internet Requirements for Local AI Model Deployment

By

–

11 May 2024 13h44

That should NEVER even end up at tech suport You should check if user has 1) GPU to run it fast 2) internet speed to download model fast If not you should not even let them use it and just give them the cloud version instead

Client-Side AI Execution: GPU and Model Size Constraints

By

–

11 May 2024 13h42

Essentially there's nothing you can NOT run client-side now The only limit is GPU size and model size to download

Distributing 70B Model Files: Infrastructure Challenges

By

–

11 May 2024 13h38

Yep but there's 70B too, issue is how do you get that 40GB file to the user

Running Llama 3 Locally in Browser with WebGPU GPU

By

–

11 May 2024 13h35

I got Llama 3 running in my browser using only my GPU with my Wi-Fi switched OFF completely client-side

WebGPU is a new feature in browsers where JS can use the GPU of the device and apparently you can run LLMs on it too, and it's fast!

My end goal for 🧠… pic.twitter.com/iz1oxeUQNp
— @levelsio (@levelsio) 11 mai 2024

I got Llama 3 running in my browser using only my GPU with my Wi-Fi switched OFF completely client-side WebGPU is a new feature in browsers where JS can use the GPU of the device and apparently you can run LLMs on it too, and it's fast! My end goal for