Prefill Decode

Prefix-cache hit lands on a request: does speculative decoding still help that request?

When does splitting prefill and decode onto separate replica pools actually pay off in production?

Why does prefill saturate compute while decode is bottlenecked on memory bandwidth?

Which workload trace shows the strongest case for disaggregated prefill/decode serving?

How does Sarathi-Serve's chunked prefill change the latency profile of a serving system?