Hi, I'm Stephen Bonner — an ML researcher at AstraZeneca working on LLMs for science, GPU performance, and Linux-based tooling. This blog covers experiments, deep-dives, and the occasional failure.
TL;DR: I was able to run local LLMs using my Intel Arc A750 GPU on a system without ReBAR by using the Vulkan backend. Updates 15 March 2026 FP16 support is now working for me on my Intel ...
TLDR. This post details how to setup a cache drive using LVM on Arch Linux. I have long been a fan of Intel’s (sadly now dead) Optane technology. Optane was pitched as a new layer sitting in b...
As I touched on in my previous post, I have been investigating local LLM offerings on Intel GPUs. As part of this, I had to gain familiarity with different nomenclature to what I was used having hi...