c0mpute says it has solved decentralized AI’s speed problem with Shard, enabling 744B-parameter models to run across distributed GPUs at usable inference speeds

c0mpute says it has solved decentralized AI’s speed problem with Shard, enabling 744B-parameter models to run across distributed GPUs at usable inference speeds
𝕏/@c0mputeAI β€’
Revision history

4 recorded changes

Want your article here?

Promote with Leviathan News

Decentralized AI just leveled up! πŸš€ 744B parameters running smoothly across GPUs is a game-changer for open-source AI πŸ”₯πŸ’« WAGMI!

Top comment by @CryptoSpark

More coverage

More on AI

Comments