Kalavai

Humble beginnings... This demo shows how to join a public cloud (our community devices!), and access shared GPUs to deploy a 3B LLM, all from the comfort of my not-so-good laptop!

PS: the model is Qwen2.5-3B-Instruct

We are ready to start testing public pools, which let users create and join computing pools to share resources (GPUs, CPUs, memory) for AI workloads.

Are you an early adopter? Join us in our community, or read more here

In a world where big tech hoards GPUs, researchers, hobbyists and open source enthusiasts have more power -together- than they think.

We believe it.

Read more: https://kalavainet.substack.com/p/a-social-network-for-ai-computation

Models are growing faster than the rate of hardware improvement, so the way to catch up is by throwing many (really many!) GPUs to the problem. And things are only going to get worse. Whereas some time ago one might have frowned at the prospect of distributed computing, today it is a feature of the AI landscape. In this new paradigm it is important to widen the doors to foster innovation.

That's our mission at Kalavai.

Read more: https://kalavainet.substack.com/p/ai-is-distributed-and-its-about-to

Kalavai, a platform that helps devs and researchers run AI by crowdsourcing computation, is now fully open source! Check it out for free: https://github.com/kalavai-net/kalavai-client

Why did we do it?

Because we want a wider audience and we are in the business of removing barriers to access AI. Making the platform open source and free is the fastest and most effective way to do both. The success of Kalavai is only possible with the support of the community that we serve.