# llama.cpp vb11391 adds CUDA tweak and new pre‑built binaries

> The open‑source LLM runner releases version b11391 with a small CUDA code change and fresh binaries for macOS, iOS and several Ubuntu platforms.

Oossa · 2026-10-04 · https://oossa.com/en/llama-cpp-vb11391-adds-cuda-tweak-and-new-pre-built-binaries

The ggml‑org team released llama.cpp version b11391 on 4 October 2026. The update moves a CUDA variable called blocks_per_col to where it is used, a change signed off by Adrien Gallouët. The release also bundles new pre‑compiled binaries for macOS Apple Silicon, macOS Intel, iOS, and multiple Ubuntu builds (CPU, Vulkan and CUDA 12). All files are available for download from the GitHub release page.

## The facts

- Version b11391 released on 2026-10-04
- Adds CUDA code change and binaries for macOS, iOS and Ubuntu (CPU, Vulkan, CUDA 12)

## Why it matters

Developers can now run llama.cpp on more hardware out‑of‑the‑box, especially those using CUDA‑enabled GPUs.

## Sources & references

1. [ggml-org/llama.cpp b11391](https://github.com/ggml-org/llama.cpp/releases/tag/b11391) – llama.cpp, 2026-10-04

Last updated: 2026-10-04