Models & Labs
New b8998 Release for Llama.cpp
llama.cpp Releasesmedium confidence
Why it matters
- →This release enhances cross-platform compatibility for developers. • It supports a wide range of architectures, improving accessibility. • The inclusion of GPU support options allows for better performance in AI applications.
The latest release of Llama.cpp, version b8998, has been announced, providing support for multiple operating systems. This update includes compatibility for macOS Apple Silicon, Intel, various Linux distributions, Android, and Windows with specific configurations for CPU and GPU support. Notably, it features enhancements for CUDA, Vulkan, and SYCL across different architectures. This release expands the accessibility of Llama.cpp for developers working on diverse 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 Releases Update for Multiple Platforms
llama.cpp Releases · April 30, 2026 · Same story
What happened next
- 2b9004 Release for Multiple Platforms
llama.cpp Releases · May 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.
