llama.cpp Update: DFlash Output Head Sharing Fixes.

Core summary

One sentence to understand this update

llama.cpp released version b11512, addressing issues such as DFlash output head sharing and enhancing GGUF metadata reading and word embedding metadata sharing.

Impact & opportunity

What this could mean

Builders will benefit from more stable DFlash functionality for llama models and improved GGUF metadata handling, enhancing the reliability of local model deployments.