web-llm icon indicating copy to clipboard operation
web-llm copied to clipboard

Decentralized distributed Supercomputing AI

Open rugabunda opened this issue 1 year ago • 2 comments

We should also integrate an option for distributed computing, like folding@home, dedicating a certain percentage of cpu/gpu power for a decentralized service, as decentralized supercomputing would significantly aid with computationally intensive AI tasks and or building language models.

rugabunda avatar Apr 20 '23 21:04 rugabunda

yes, with webgpu we can train llm with many many chrome on small computers,maybe need p2p and tranning tasks manager?

eatcosmos avatar Apr 21 '23 03:04 eatcosmos

I like the idea of a swarm LLM ;-).

HappyPony avatar Apr 24 '23 07:04 HappyPony