# llama.cpp patch speeds up one RTX 5090 matrix multiply 4.5x

> Source: <https://runtimewire.com/article/llama-cpp-rtx-5090-q4k-n16-vulkan-4-5x>
> Published: 2026-11-15 01:43:49+00:00

# llama.cpp patch speeds up one RTX 5090 matrix multiply 4.5x

**Jeff Bolz's narrow tiles cut the KHR cooperative-matrix case at M=4096, K=14336, N=16 from 325.49 to 72.36 microseconds on an RTX 5090.**

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

llama.cpp, the inference project created by Georgi Gerganov, merged new Vulkan cooperative-matrix tiles on November 14th, 2026. The sharpest result was for one matrix shape: Q4_K, M=4096, K=14336, N=16, on an NVIDIA GeForce RTX 5090. In the KHR cooperative-matrix path, that multiply fell from 325.49 microseconds to 72.36 microseconds, a 4.50x kernel speedup reported in the merged pull request.…