Open Source
llama.cpp b10159 release enhances Metal backend
llama.cpp Releaseshigh confidence
Why it matters
- →The new FWHT kernel improves performance for Apple Silicon users.
- →Fixing the narrowing issue enhances stability and reliability.
- →Broad platform support ensures versatility in diverse environments.
The b10159 release of llama.cpp brings a new FWHT kernel to the Metal backend, enhancing performance for Apple Silicon devices. Co-authored by YiChen Lv and Georgi Gerganov, the update also fixes a narrowing issue and includes minor formatting improvements. Despite the KleidiAI feature being disabled for macOS Apple Silicon, the release supports multiple platforms like Ubuntu, Windows, and openEuler. This update solidifies llama.cpp's role as a flexible inference tool across various hardware setups.
Read originalThe story around this
Earlier coverage that leads up to this article, and what followed. Lines connect each piece to the closest one after it, converging here.
What happened next
- 1llama.cpp b10232 release enhances Metal support
llama.cpp Releases · August 3, 2026 · Same story
- 2llama.cpp b11059 optimizes Metal FWHT for Apple Silicon
llama.cpp Releases · September 20, 2026 · Same story
More from llama.cpp Releases
Coding Toolscoding
llama.cpp b12022 adds ROCm 10 and Snapdragon support
This release quietly cements llama.cpp as the universal inference runtime by finally bringing ROCm 10.0 to Linux and Windows alongside CUDA 13.4, effectively closing the hardware gap for AMD users who were previously left behind. The inclusion of native Snapdragon support on Linux marks a significant step toward running local LLMs on ARM-based mobile chips without emulation overhead. While KleidiAI on Apple Silicon is disabled in this build, the broader expansion to diverse accelerators ensures that developers aren't locked into NVIDIA ecosystems anymore.
llama.cpp Releases
Coding Toolscoding
llama.cpp b12023 fixes shared prefix bug
This release addresses a specific inference artifact where decision prefixes were incorrectly bleeding into output embeddings. By capping shared prefixes at the first output position, llama.cpp ensures that option tokens remain distinct for parent and child tasks. This is a targeted fix for multi-step reasoning or voting workflows rather than a performance boost. It stabilizes complex generation patterns without changing the underlying model architecture.
