Tech blog
  • Blog
  • Github
  • Twitter
  • luigi@grigio.org

strata

A collection of 1 post
strata

Strata vulkan native expert streaming vs llama.cpp Vulkan on an Intel Arc B390 iGPU

Measured 19 November 2026 on one machine, one model, one day. Short version: on an Intel Arc B390 iGPU, llama.cpp Vulkan is about 14× faster at decode and batches a realistic prompt about 870× faster at prefill, but Strata's native streaming engine keeps the footprint at about
19 Nov 2026 13 min read
Page 1 of 1
Tech blog © 2026
  • Sign up
Powered by Ghost