Author

Omkar Shewale

1 paper indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

#machine learning Preprint Sep 2026

PrefixBench-H100: Characterizing Prefix Reuse and Time-to-First-Token in H100 LLM Serving

The goal of PrefixBench-H100 is to expose the practical operating envelope of prefix reuse for H100-class LLM serving, and identifies the regime where prefix reuse provides substantial first-token latency reductions and the regime where cache pressure erodes them.

Omkar Shewale, Deepak Kumar, Divakar Yadav · 0 citations