Models & Labs
llama.cpp b10990 Release Expands Platform Support
llama.cpp Releaseshigh confidence
Why it matters
- →Expands support for diverse hardware, including non-NVIDIA systems.
- →Enhances compatibility with new libraries like ROCm 10.0.
- →Facilitates broader AI model deployment across various platforms.
The b10990 release of llama.cpp has been announced, featuring expanded support for various platforms and architectures. Notably, it includes ROCm 10.0 support for both Ubuntu and Windows, alongside updates for Vulkan and CUDA libraries. While no new models are introduced, the release enhances compatibility across Apple Silicon, Intel, and multiple Linux distributions. This update aims to improve the deployment of AI models on a broader range of hardware, making it more accessible for developers using non-NVIDIA systems.
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 b10901 Release Expands Platform Support
llama.cpp Releases · September 11, 2026 · Same story
- 2llama.cpp b10992 Release Expands Platform Support
llama.cpp Releases · September 16, 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.
