github llama.cpp · 3d ago · inference

llama.cpp b10830 adds --fuse-qkv flag to HF-to-GGUF conversion

from English

The llama.cpp project has released version b10830, introducing a new command-line option for model conversion. This update adds the `--fuse-qkv` flag, which allows users to fuse Query, Key, and Value matrices into a single QKV structure during the conversion from Hugging Face format to GGUF.

The release includes binaries for macOS (Apple Silicon and Intel), Linux (Ubuntu with CPU, Vulkan, ROCm, OpenVINO, and SYCL backends), Android, Windows (CPU, CUDA 12/13, Vulkan, OpenCL, ROCm, OpenVINO, and SYCL), and openEuler.
iOS support is provided via an XCFramework, while KleidiAI on macOS Apple Silicon is currently disabled.

This update provides users with a more efficient conversion path for models that benefit from fused attention mechanisms.

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

Read original

github llama.cpp · 2d ago · 2 views

llama.cpp disables lazy tensor loading by default on iGPUs

The llama.cpp project has disabled lazy tensor loading by default for integrated GPUs (iGPUs) in release b10867. This change addresses a regression introduced in the previous "AUTO" lazy mode setting.

github llama.cpp · 2d ago · 3 views

llama.cpp b10864 fixes checkpoint eviction for hybrid models

The llama.cpp project released build b10864, which includes a fix for the checkpoint spacing eviction logic in `create_checkpoint()`. Previously, the eviction rule kept the oldest checkpoint and erased later ones within `checkpoint_min_step`, causing prompts shorter than that threshold to drop the checkpoint needed for resumption. This forced hybrid and recurrent models to unnecessarily re-prefill from the previous checkpoint.

github llama.cpp · 3d ago · 6 views

llama.cpp b10844 adds Vulkan DeepSeek-V4 hyper-connection fused ops

The llama.cpp project released version b10844, introducing support for DeepSeek-V4 hyper-connection fused operations (DSV4_HC_COMB/PRE/POST) in the Vulkan backend. This update brings Vulkan to parity with CUDA and Metal regarding these specific optimizations.

github llama.cpp · 3d ago · 4 views

llama.cpp b10842 adds gfx90c HIP support

The llama.cpp project released build b10842, which introduces support for the gfx90c architecture via HIP.

media Hugging Face Forums · 3d ago · 8 views

AI Model Radar launches with measured hardware fit for open models

A new site called AI Model Radar helps users determine which open models actually fit their GPU or Mac by using measured data rather than estimates. The platform calculates weight sizes based on real published GGUF files and computes context cache layer-by-layer from model configurations.