# llama.cpp version b11537 released with embedding fixes and new binaries

> The ggml-org/llama.cpp library adds graph reorder for embeddings, fixes for Gemma4, and ships pre‑built binaries for macOS, iOS and Linux.

Oossa · 2026-10-09 · https://oossa.com/en/llama-cpp-version-b11537-released-with-embedding-fixes-and-new-binaries

ggml-org released llama.cpp version b11537. The update reorders graph get_rows for embeddings and fixes Gemma4 embedding handling. It also adds a TODO note for LoRA support and corrects raw embedding paths. Pre‑compiled binaries are now available for macOS Apple Silicon, macOS Intel, iOS, and various Ubuntu builds, including Vulkan‑enabled versions.

## The facts

- Version b11537 released on 2026-10-09
- Binaries provided for macOS (arm64 and x64), iOS, and Ubuntu (x64, arm64, s390x, Vulkan)

## Why it matters

Developers can now run llama.cpp locally on more platforms with corrected embedding support, reducing setup hassle.

## Sources & references

1. [ggml-org/llama.cpp b11537](https://github.com/ggml-org/llama.cpp/releases/tag/b11537) – llama.cpp, 2026-10-09
2. [ggml-org/llama.cpp b11538](https://github.com/ggml-org/llama.cpp/releases/tag/b11538) – llama.cpp, 2026-10-09

Last updated: 2026-10-09