Qeravio
Canonical AI event

CUDA 12.8 libraries added for Ubuntu x64.

Sparse FA for qwen4 is now enabled on CUDA 13.3 libraries for Ubuntu arm64.

20 Sept 20261 verified claims1 sources1 observations
What happened

The official source reports this update: b11062: CUDA: enable sparse fa for qwen4.

Why it matters

This official update documents a development concerning b11062: CUDA: enable sparse fa for qwen4. 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
infrastructureMamba time-step projection input now contiguous in llama.cpp20 Sept 2026infrastructureHexagon Adds TOP_K Operation Support20 Sept 2026infrastructureHexagon adds GEGLU_QUICK support across multiple platforms.20 Sept 2026infrastructureLlama.cpp adds I32 GET_ROWS support across multiple platforms20 Sept 2026infrastructureLlama.cpp server logs now show source per model.20 Sept 2026infrastructureJson-schema now supports escaped hyphens in regex patterns.20 Sept 2026
Evidence trail

Sources behind the event

llama.cpp official releasesb11062: CUDA: enable sparse fa for qwen4
Editorial presentation

Read the public story

Open story