github llama.cpp · just now · inference

llama.cpp b10721 fixes WebGPU crash and adds Windows CUDA 13.4 support

from English

The llama.cpp project has released build b10721, which includes a critical fix for the WebGPU backend and expands hardware support on Windows.

Fixes a crash in the WebGPU `ggml_backend_tensor_get()` implementation when the offset is not a multiple of 4.
Adds preview support for Windows arm64 with CUDA 13.4.
Provides binaries for macOS (Apple Silicon and Intel), Linux (CPU, Vulkan, ROCm, OpenVINO, SYCL), Android, and Windows across various CPU and GPU backends.

This release ensures stability for WebGPU users on non-aligned memory offsets and extends compatibility to newer CUDA architectures on Windows.

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

Read original

github llama.cpp · 2h ago

llama.cpp b10718 extends MOE fusion to specdec and multi-token

The llama.cpp project released version b10718, which includes a significant update to its CUDA backend. The most notable change is the extension of Mixture of Experts (MOE) fusion to support speculative decoding (specdec), addressing previous limitations where MOE glu fusion and topk-router fusion were restricted to processing only one token at a time.

github llama.cpp · 5h ago · 1 view

llama.cpp b10715 fuses DFlash encoder into KV cache injection

The llama.cpp project released build b10715, which includes a key optimization to the DFlash implementation. The DFlash encoder is now fused directly into the KV cache injection process during decoding.

github llama.cpp · 6h ago · 1 view

llama.cpp b10714 tunes Vulkan mat-vec rows for Strix Halo

The llama.cpp project released build b10714, which includes specific optimizations for the Vulkan backend targeting AMD Strix Halo hardware. The update adjusts matrix-vector multiplication row counts to improve performance during batched inference.

github llama.cpp · 12h ago · 3 views

llama.cpp b10706 adds SWIGLU_CLAMP and Vulkan shader support

The llama.cpp project has released build b10706, introducing new functionality to the ggml backend. This update includes the addition of SWIGLU_CLAMP and a new Vulkan shader implementation.

github llama.cpp · 13h ago · 9 views

llama.cpp b10707 optimizes KV cache cell scanning for faster generation

The llama.cpp project released version b10707, which includes a performance optimization to the key-value (KV) cache handling. The update modifies the `for_each_token_in` function to stop scanning once all sequences within a specific cell have been seen, rather than iterating through all possible maximum sequences.