Compute-Optimal Is Not Cluster-Optimal

(szha.ai)

2 points | by yuxin_tang 6 hours ago ago

1 comments

  • yuxin_tang 6 hours ago

    https://arxiv.org/pdf/2608.10605: Our new paper folds the systems stage into the scaling-law stage. Price every candidate architecture on what the cluster actually delivers, and the answer changes: the sparsity an MoE 'should' have depends on the cluster you train it on.