Decentralized training already works at 405B scale
Akash Accelerate 2025 - Greg Osuri Keynote (Akash Network)
“Nous pushed scale even further — a 405 billion parameter model trained in the wild, fully open source, fully decentralized. This isn’t the future. It’s the present. Together, these breakthroughs prove… we can train powerful models without centralized servers. We can verify correctness without trust. And we can scale intelligence without monopolies.” — 00:07:17
Context: After citing Pluralis (swarm training), Gensyn (72B on-chain), and Bagel (ZK-LoRA) as existence proofs of decentralized AI.