ggml-org/llama.cpp b8238

on GitHub

llama: end-to-end tests (#19802)

• tests: add end-to-end tests per model architecture

• fixup for rebase

• fix use-after-free in llama-model-loader.cpp

• fix CI

• fix WebGPU

• fix CI

• disable CI for macOS-latest-cmake-arm64

• use expert_weights_scale only if != 0.0f

• comments