Models & Labs
Llama.cpp b9255 Release Enhances HMX Matmul
llama.cpp Releaseshigh confidence
Why it matters
- →Enhances performance of AI computations with optimized matrix multiplication.
- →Streamlines configuration by combining power and clock settings.
- →Improves compatibility with Snapdragon and Apple Silicon devices.
Llama.cpp has released its b9255 update, which brings significant improvements to the HMX quantized matrix multiplication. The update refines the dequantization logic and removes the non-pipelined version to enhance performance. It also includes updates for Snapdragon devices and Apple Silicon, aiming to streamline AI processing across platforms. These changes are part of ongoing efforts to optimize AI computations and improve compatibility with various hardware configurations.
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 b9387 Release Enhances AMD MFMA Performance
llama.cpp Releases · May 29, 2026 · Same story
- 2Llama.cpp b9928 Release Enhances VTCM and HMX Pipelines
llama.cpp Releases · July 9, 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.
