Press Space to continue
K3 Achieves 947 Tokens Per Second Decoding on Single GPU
Press Space to continue
single node decoding speed for K3 is actually pretty fast considering it's a 2.8T model on a single B300 node 947 tokens/s decoding throughput at batch size 32 and you get even get 152 tokens/s for a single user
open on x · 2 viewsViews and clicks count once per visitor per day. Clicks are opens of this post on X.