llama.cpp b11009 Update Fixes QKV Splitting for Gemma4 and Qwen35

Core summary

One sentence to understand this update

llama.cpp has released update b11009, which includes a fix for split state and granularity issues in fused QKV for Gemma4 and Qwen35 models.

Impact & opportunity

What this could mean

Developers working with Gemma4 and Qwen35 models in llama.cpp will benefit from improved stability and correctness, enabling more reliable local deployments and experiments.