# llama.cpp b11895 fixes a Metal offset overflow

> The ggml-org release widens offsets and token indices in Metal matrix operations, addressing overflow in large matrix outputs.

Oossa · 2026-10-26 · https://oossa.com/en/llama-cpp-b11895-fixes-a-metal-offset-overflow

ggml-org released llama.cpp b11895 with a fix for overflow in large matrix output offsets on Metal. The change also widens batch offsets and token indices used by the MUL_MAT_ID operation.

The release page provides downloads for Apple Silicon and Intel Macs, iOS, and several Ubuntu CPU and Vulkan builds.

## The facts

- The Metal fix is listed as change #30072.
- The release includes macOS Apple Silicon and Intel binaries, plus an iOS XCFramework.

## Why it matters

Developers running llama.cpp on Metal may avoid offset overflow in large matrix outputs, though the release notes do not say how often the issue occurred or which workloads trigger it.

## Sources & references

1. [ggml-org/llama.cpp b11895](https://github.com/ggml-org/llama.cpp/releases/tag/b11895) – llama.cpp, 2026-10-26

Last updated: 2026-10-26