# llama.cpp b11971 adds tiled OpenCL support for Adreno

> The release adds a tiled q4_K matrix operation for Adreno GPUs, including a path for batches larger than one.

Oossa · 2026-10-28 · https://oossa.com/en/llama-cpp-b11971-adds-tiled-opencl-support-for-adreno

ggml-org’s llama.cpp b11971 release adds a tiled OpenCL q4_K matrix operation for Adreno GPUs. The change also routes the q4_K language-model output layer through that operation when the batch size is greater than one.

The release includes downloadable builds for macOS, iOS and several Linux platforms. The release notes do not give performance measurements, so the speed or resource impact of the Adreno change is unclear.

## The facts

- The Adreno change is listed as pull request #30412.
- The release provides macOS Apple Silicon and Intel builds, an iOS package, and Linux builds.

## Why it matters

Developers running llama.cpp on Adreno hardware can try the new OpenCL path, but the release notes do not say whether it makes inference faster.

## Sources & references

1. [ggml-org/llama.cpp b11971](https://github.com/ggml-org/llama.cpp/releases/tag/b11971) – llama.cpp, 2026-10-28

Last updated: 2026-10-28