# llama.cpp release b11268 adds OpenCL fix for Adreno GPUs

> The open‑source LLM runner updates its OpenCL code to correct tensor handling on Adreno GPUs and ships new binaries for macOS, iOS and Linux.

Oossa · 2026-09-30 · https://oossa.com/en/llama-cpp-release-b11268-adds-opencl-fix-for-adreno-gpus

The ggml‑org team released llama.cpp version b11268. The update fixes a bug in the OpenCL driver that handled tensors for the q5_K Adreno GPU kernel. New pre‑built binaries are provided for macOS Apple Silicon, Intel Macs, iOS, and several Linux configurations including CPU, Vulkan and CUDA 12.8.

## The facts

- Release tag b11268 published Sep 30, 2026
- Fix targets OpenCL q5_K adreno gemm_nonshuffle kernel

## Why it matters

The fix improves performance and stability for developers running LLMs on Android devices with Adreno GPUs.

## Sources & references

1. [ggml-org/llama.cpp b11268](https://github.com/ggml-org/llama.cpp/releases/tag/b11268) – llama.cpp, 2026-09-30

Last updated: 2026-09-30