github llama.cpp · 1d ago · inference

llama.cpp b10871 adds dedicated iq4_xs mat-vec shader for Vulkan

from English

The llama.cpp project released build b10871, introducing a dedicated mul_mat_vec_iq4_xs shader for the dmmv path in Vulkan. This change replaces the previous generic fallback implementation to improve performance on RDNA4 hardware.

The new shader provides approximately 6-17% faster token generation depending on the model used.
An experimental branch attempting to fully unroll the block loop for n_it <= 8 was removed as it was functionally equivalent to the simple loop kept in the code.
Binaries are available for macOS, Linux, Windows, Android, and openEuler across CPU, CUDA, ROCm, OpenVINO, SYCL, and Vulkan backends.

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

Read original

github llama.cpp · 2h ago · 1 view

llama.cpp b10888 adds Vulkan command-buffer debug labels for GPU profilers

The llama.cpp project has released build b10888, which includes a new feature to add command-buffer debug labels for GPU profilers in the Vulkan backend.

github llama.cpp · 3h ago · 1 view

llama.cpp b10886 adds Q1_0 vector intrinsic support for s390x

The llama.cpp project released build b10886, which introduces Q1_0 vector intrinsic support for the s390x architecture. This update includes the implementation of `ggml_vec_dot_q1_0_q8_0` and updates documentation to reflect the new capability.

github llama.cpp · 17h ago · 1 view

llama.cpp b10883 fixes Vulkan matrix multiplication and restores Ampere optimizations

The llama.cpp project released build b10883, which includes significant updates to the Vulkan backend. The primary change involves using spec constants for matrix-matrix multiplication A-type operations to improve shader compilation and performance.

github llama.cpp · 1d ago · 2 views

llama.cpp b10870 adds f16 B-type matmul pipelines and warp tile tuning for Intel coopmat1

The llama.cpp project released version b10870, introducing Vulkan support for f16 B-type matrix multiplication pipelines and warp tile size tuning specifically for Intel coopmat1 hardware.

github llama.cpp · 2d ago · 1 view

llama.cpp b10863 fixes idle threads in Metal mul_mv_iq3_xxs for small matrices

The llama.cpp project released build b10863, which includes a fix for idle threads in the `mul_mv_iq3_xxs` kernel on Metal when `ne00` is less than 1024. The update addresses a half-idle simdgroup issue by introducing a separate split kernel for specific matrix dimensions.