# llama.cpp adds new Q8_0 and i8mm kernels for Apple and Linux

> The open‑source LLaMA inference library now includes faster quantized kernels for ARM SVE and i8mm, plus new binaries for macOS, iOS and Ubuntu.

Oossa · 2026-10-22 · https://oossa.com/en/llama-cpp-adds-new-q8-0-and-i8mm-kernels-for-apple-and-linux

The llama.cpp project released version b11811 on Oct 22 2026. The update adds Q8_0 SVE dot‑product GEMV and SVE i8mm GEMM kernels, which speed up quantized model inference on ARM CPUs. Pre‑built binaries for macOS Apple Silicon, Intel, iOS, and several Ubuntu builds (CPU, Vulkan, CUDA 12.8) are now available for download.

## The facts

- Release version b11811 published Oct 22 2026
- New kernels: Q8_0 SVE dot‑product GEMV and SVE i8mm GEMM

## Why it matters

Developers can run LLaMA models faster on newer ARM hardware and on Linux GPUs without writing custom code.

## Sources & references

1. [ggml-org/llama.cpp b11811](https://github.com/ggml-org/llama.cpp/releases/tag/b11811) – llama.cpp, 2026-10-22

Last updated: 2026-10-22