Models & Labs
llama.cpp b10545 Release Fixes Tensor API Bug
llama.cpp Releaseshigh confidence
Why it matters
- →Fixes a critical bug in tensor operations, preventing data corruption.
- →Enhances reliability for developers using the Tensor API.
- →Ensures accurate matrix computations even with unaligned K inputs.
The llama.cpp b10545 release resolves a significant issue in the Tensor API's mat-mat kernel, which previously caused undefined behavior when the K dimension wasn't a multiple of 32. By implementing a dynamic extent for K, the update ensures that the kernel only processes valid data, preventing potential corruption or NaN results. This change is particularly important for developers working with unaligned K inputs, as it improves the accuracy and reliability of tensor operations.
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 b9275 Release Optimizes Metal Kernels
llama.cpp Releases · May 22, 2026 · Same story
What happened next
- 2llama.cpp b11109 fixes precision bugs
llama.cpp Releases · September 22, 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.
