Models & Labs
Llama.cpp b9244 Release Adds MoE Support
llama.cpp Releaseshigh confidence
Why it matters
- →MoE support on Adreno GPUs enhances OpenCL's versatility for AI tasks.
- →Broad platform compatibility ensures developers can use llama.cpp across diverse systems.
- →The update strengthens llama.cpp's position as a flexible tool for AI development.
The b9244 release of llama.cpp brings MoE support for q4_k, q5_k, and q6_k on Adreno GPUs, enhancing its OpenCL capabilities. This update, co-authored by Li He from Qualcomm, expands the tool's functionality for developers using Qualcomm hardware. The release also includes builds for macOS Apple Silicon, Windows with CUDA, and various Linux configurations, ensuring broad compatibility. These improvements make llama.cpp a more versatile option for AI developers working across different platforms.
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.
How we got here
- 1Llama.cpp b9142 Release Expands OpenCL Support
llama.cpp Releases · May 14, 2026 · Same story
What happened next
- 2llama.cpp b9852 release expands OpenCL support
llama.cpp Releases · July 2, 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.
