llama.cpp b9565 Release Enhances ggml-webgpu

Core summary

One sentence to understand this update

The llama.cpp project released version b9565, which primarily focuses on enhancements for ggml-webgpu, including handling buffer overlap/aliasing for the concat operator and optimizing WebGPU CI.

Impact & opportunity

What this could mean

Developers using llama.cpp for WebGPU deployments will benefit from more stable concat operations and optimized build processes, improving model execution efficiency.