github llama.cpp · 2d ago · inference

llama.cpp b10937 applies noshuffle row-alignment rule to q4_K, q5_K and q8_0

from English

The llama.cpp project released version b10937, which extends the OpenCL noshuffle row-alignment rule to q4_K, q5_K, and q8_0 quantization formats. Previously, this alignment rule was applied only to q6_K.

The change addresses OpenCL compatibility for specific quantization types.
Binaries are provided for macOS (Apple Silicon and Intel), Linux (CPU, Vulkan, ROCm, OpenVINO, SYCL), Android, Windows (CPU, CUDA, Vulkan, OpenVINO, SYCL, ROCm), and openEuler.
An iOS XCFramework and UI build are also included in the release.

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

Read original

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

llama.cpp b10984 adds CUDA row-contiguous SUM_ROWS support

The llama.cpp project released version b10984, introducing CUDA support for row-contiguous SUM_ROWS operations. This update also organizes the code to include GGML_OP_MEAN for handling row-contiguous tensors via a shared kernel.

github llama.cpp · 2h ago · 1 view

llama.cpp b10985 restricts RPC hash-cache to weights

The llama.cpp project released build b10985, which modifies the RPC backend to limit hash-based caching exclusively to weight tensors. This change addresses an issue where activation data was unnecessarily hashed and stored, leading to excessive disk usage.

github llama.cpp · 4h ago · 2 views

llama.cpp b10981 fixes stateful decode for Gemma-4 and optimizes GPU MoE inference

The llama.cpp b10981 release addresses critical correctness issues in the OpenVINO backend's stateful decoding path, specifically fixing support for models like Gemma-4 that use per-layer head sizes. It also introduces optimizations for GPU Mixture of Experts (MoE) inference and improves performance through KV state relayouts.

github llama.cpp · 5h ago · 1 view

llama.cpp adds Metal FA kernels for MiniCPM3's HSK=96, HSV=64

The llama.cpp project has added Flash Attention (FA) kernel support for the specific configuration used by MiniCPM3 on Apple Silicon. This change addresses a crash where the `-fa auto` flag would abort due to a missing kernel instantiation for attention key length 96 and value length 64.

github llama.cpp · 19h ago · 3 views

llama.cpp b10970 release uses fp32 accumulation on MFMA devices

The llama.cpp project has released version b10970, which includes a technical update to the fattn-mma backend. This change implements the use of fp32 accumulators specifically for CDNA architecture devices.