Qeravio
Canonical AI event

Llama.cpp adds A8 Q6_K non-MoE dp4a binary kernel

The update includes support for Ubuntu s390x and Windows arm64 OpenCL Adreno.

24 Sept 20261 verified claims1 sources1 observations
What happened

The official source reports this update: b11147: opencl: add A8 Q6_K non-MoE dp4a binary kernel.

Why it matters

This official update documents a development concerning b11147: opencl: add A8 Q6_K non-MoE dp4a binary kernel. Its practical significance depends on the scope and evidence stated by the source.

What to watch next

Read the official source update and verify its stated scope, evidence and timing before acting on it.

Connected knowledge

Entities affected by this event

Llama Cppsubject
Continue this topic
infrastructureGemma 4 draft support added to DSpark with GGUF conversion24 Sept 2026infrastructureLlama.cpp server fixes crash on sleep, updates wake-up process24 Sept 2026infrastructureDraft HF model deduplication implemented via dedup-cache-models24 Sept 2026infrastructureCUDA sparse-fa for dsv4 prefill re-enabled, query loop unrolled.24 Sept 2026infrastructurellama.cpp updated to version 0.5.0, now supports CUDA 13.4 libraries.24 Sept 2026infrastructureVulkan adds IQ4_XS MMQ/MMV matmul kernels, disables MMVQ on Intel.24 Sept 2026
Evidence trail

Sources behind the event

llama.cpp official releasesb11147: opencl: add A8 Q6_K non-MoE dp4a binary kernel
Editorial presentation

Read the public story

Open story