# llama.cpp v0.1.15 adds CUDA out‑of‑bounds fix and new binaries

> The ggml‑org release on Oct 8, 2026 patches a CUDA memory issue and provides pre‑built binaries for macOS, iOS and several Linux platforms.

Oossa · 2026-10-08 · https://oossa.com/en/llama-cpp-v0-1-15-adds-cuda-out-of-bounds-fix-and-new-binaries

The open‑source llama.cpp library released version b11511 on Oct 8, 2026. It fixes a CUDA out‑of‑bounds read bug in the MMQ implementation. The update also adds ready‑to‑run binaries for macOS (Apple Silicon and Intel), iOS, and Ubuntu on x64, arm64 and s390x, plus Vulkan and CUDA‑12.8 builds.

## The facts

- Release tag: b11511
- Fix: CUDA MMQ out‑of‑bounds reads (#29953) – released 2026‑10‑08

## Why it matters

Developers can now run llama.cpp on more hardware without crashes, especially those using NVIDIA GPUs.

## Sources & references

1. [ggml-org/llama.cpp b11511](https://github.com/ggml-org/llama.cpp/releases/tag/b11511) – llama.cpp, 2026-10-08

Last updated: 2026-10-08