Open Source
Llama.cpp Adds Ubuntu-CUDA Builds in Latest Update
llama.cpp Releaseshigh confidence
Why it matters
- →Expands compatibility with CUDA on Ubuntu, benefiting developers using these systems.
- →Enhances infrastructure with GCC 14 support for modern development environments.
- →Addresses licensing and dependency management, improving usability and compliance.
Llama.cpp has released a new update, b10969, which includes the addition of Ubuntu-CUDA build jobs for versions 12.8 and 13.3 on x64 and arm64 architectures. This update also incorporates GCC 14 for CUDA arm64 builds, enhancing compatibility and performance for developers using these systems. The update addresses infrastructure improvements such as shipping dependent libraries on Ubuntu and managing NCCL licensing. These changes aim to streamline the development process for those utilizing CUDA on Ubuntu, making llama.cpp a more versatile tool.
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 version 0.3.0
llama.cpp Releases · August 26, 2026 · Same story
What happened next
- 2llama.cpp b10964 release expands platform support
llama.cpp Releases · September 15, 2026 · Same story
Follow this story
Open the full 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.
