# Tensward finds 643 MB copied off an H100 for each request

> Source: <https://runtimewire.com/article/tensward-gpu-census-643-mb-h100>
> Published: 2026-11-15 18:42:39+00:00

# Tensward finds 643 MB copied off an H100 for each request

**Tensward, founded by Adi Shik, traced the transfers in one H100 test to llama.cpp state saving.**

        By [RuntimeWire Staff](https://runtimewire.com/author/runtimewire-staff)
        · Published 

Adi Shik started Tensward after a client deployment turned into a lesson in inference optimisation. He has described trying to deploy a local model in a client's environment in a previous job, calling inference optimisation "a huge headache" and saying the experience gave him the idea for Tensward.

In Tensward's November 15th post, Shik's team put an LLM server on an NVIDIA H100, pointed sixteen clients at it, and watched 'nvidia-smi' report 35% busy. Throughput was 58 tokens per second, and users waited almost two seconds for a first token.…