Models & Labs
Llama.cpp b9222 Release Adds Hexagon TRI Support
llama.cpp Releaseshigh confidence
Why it matters
- →Enhances support for Qualcomm's Hexagon DSP, improving performance for specific hardware users.
- →Resolves previous integration issues, ensuring smoother operation across platforms.
- →Strengthens llama.cpp's position as a versatile inference runtime without introducing new models.
The b9222 release of llama.cpp brings new support for the TRI HVX Kernel in ggml hexagon HTP operations. This update, co-authored by Qualcomm engineers, addresses previous review comments and resolves merge conflicts, enhancing the software's compatibility and performance. The release supports a wide range of platforms, including macOS, Linux, and Windows, but does not introduce new model architectures. This update is particularly relevant for developers utilizing Qualcomm's Hexagon DSP, as it improves the integration and functionality of llama.cpp in these environments.
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 b9082 Release Enhances Hexagon Backend
llama.cpp Releases · May 9, 2026 · Same story
What happened next
- 2Llama.cpp b9221 Release Enhances Hexagon Support
llama.cpp Releases · May 19, 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.
