github llama.cpp · 13h ago · inference

llama.cpp b10754 fixes OpenCL out-of-bound reads in Adreno image kernels

from English

The llama.cpp project released build b10754, which includes a critical fix for OpenCL on Adreno GPUs to prevent out-of-bounds memory reads in image processing kernels.

The q4_K decode GEMV fetch row is now clamped on a padded x-grid.
The tiling contract for image KQ/KQV GEMMs is strictly enforced.
The split logic for image KQ/KQV is determined at dispatch time rather than relying on strides.

This update ensures stability and correctness for users running llama.cpp with OpenCL acceleration on Adreno hardware.

Importance 1/3 Trust 1/3 llama.cpp Inference efficiency

Read original

github llama.cpp · just now Live

llama.cpp b10766 adds DeepSeek 4 vision support

The llama.cpp project has released build b10766, which includes a fix to correctly support input vision for the DeepSeek 4 model.

github llama.cpp · 1h ago · 1 view Live

llama.cpp b10762 adds DeepSeek-V4-Flash-Vision-Exp support

The llama.cpp project has released build b10762, which introduces support for the DeepSeek-V4-Flash-Vision-Exp model. This update also includes handling of minimum and maximum token counts from the command line interface.

github llama.cpp · 1h ago · 1 view Live

llama.cpp b10763 enables preserve_reasoning by default

The llama.cpp project released build b10763, which changes the default behavior of the `preserve_reasoning` chat template keyword argument. The server now logs the effective state of this kwarg and issues a warning when it is enabled by default on templates that support it.

github llama.cpp · 8h ago · 1 view

llama.cpp b10760 fixes Qwen3-tts-0.6b inference stability

The llama.cpp project released build b10760, which includes specific fixes for the Qwen3-tts-0.6B model to prevent numerical instability during inference.

github llama.cpp · 9h ago · 1 view

llama.cpp b10758 adds MUL_MAT and MUL_MAT_ID fusion for Hexagon

The llama.cpp project released build b10758, introducing MUL_MAT and MUL_MAT_ID fusion capabilities for the Hexagon DSP. This update also includes several fixes to improve stability and performance on Qualcomm hardware.