Yep, but models will get smaller and GPUs bigger esp now that AI is so popular I think
HARDWARE
-
Client-Side AI Execution: GPU and Model Size Constraints
By
–
Essentially there's nothing you can NOT run client-side now The only limit is GPU size and model size to download
-
Running Llama 3 Locally in Browser with WebGPU GPU
By
–
I got Llama 3 running in my browser using only my GPU with my Wi-Fi switched OFF completely client-side
— @levelsio (@levelsio) 11 mai 2024
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/iz1oxeUQNpI 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
-
Cerebras Engineer Discusses AI Accelerators for HPC Applications
By
–
Join Cerebras HPC Solutions Engineer, Leighton Wilson, as he speaks on a panel at ISC discussing democratizing AI accelerators for HPC applications. Date: May 14th 2024 Time: 2:15pm to 3:15pm Local Time Location: ISC 2024 in Hamburg, Germany, Hall E 2nd Floor For more
-
XPeng AeroHT Flying Car: The Future of Commuting
By
–
Your next commuting will be by a #Flying Car by @XPENG_AEROHT v/ @FrRonconi #Mobility #EVs #Innovation #Design #Aviation #AutonomousVehicles #SelfDrivingCars #electriccars #ElectricVehicles #Automotive #5G
-
Cerebras SVP discusses sustainable supercomputing design challenges
By
–
Join Cerebras SVP of Product and Strategy, Andy Hock, as he speaks on a panel at the Sustainable Supercomputing Workshop where he discussed the opportunities and challenges in designing sustainable supercomputing systems from hardware to software. Date: May 16th 2024 Time:
-
MathWorks NVIDIA Partnership Accelerates Medical AI Development
By
–
MathWorks and NVIDIA join forces to accelerate medical technology @MATLAB now integrates with NVIDIA Holoscan, enabling engineers to develop AI apps faster and more efficiently. Learn more about this partnership: http://
spr.ly/6019jagEU #NVIDIA #MATLAB #AI #MedTech -
Festo’s BionicBee: Autonomous Robotic Swarm Technology Innovation
By
–
Festo has unveiled BionicBee, a #robotic innovation that mimics bee swarming behavior autonomously. This biomimetic swarm #technology has vast potential applications across civilian and military sectors. #AI #IoT #5G @SpirosMargaris @gvalan @Hal_Good
-
Apple Silicon Mac Pro thermal efficiency improvements analyzed
By
–
I've been researching the thermal outputs of the Mac Pros and it's quite impressive the improvements due to Apple Silicon: Energy efficency, for the win (Also i'm learning about the BTU/h unit)
-
Google Gemma 10M Context Window: Local LLM Breakthrough
By
–
Google Gemma with a 10M context window LLM features:
• 1250x context length of base Gemma
• Requires less than 32GB of memory
• Infini-attention + activation compression Imagine if you can run 10 million context LLM locally on your computer with @ollama or @LMStudioAI
.