feat: cross-stage offload modes and layer-streaming for low-VRAM GPUs #2444

Job	Run time
ubuntu-latest-cmake	6m 57s
ubuntu-latest-cmake-vulkan	9m 25s
windows-latest-rocm	5m 28s
ubuntu-latest-rocm (7.13.0, gfx906;gfx908;gfx90a;gfx942;gfx950;gfx1010;gfx1011;gfx1012;gfx1030;gf...	4m 14s
macOS-latest-cmake	8m 32s
windows-latest-cmake (avx512, -DGGML_NATIVE=OFF -DGGML_AVX512=ON -DGGML_AVX=ON -DGGML_AVX2=ON -DS...	3m 41s
windows-latest-cmake (cuda12, -DSD_CUDA=ON -DSD_BUILD_SHARED_LIBS=ON -DCMAKE_CUDA_ARCHITECTURES='...	2h 4m 17s
windows-latest-cmake (avx, -DGGML_NATIVE=OFF -DGGML_AVX=ON -DGGML_AVX2=OFF -DSD_BUILD_SHARED_LIBS...	3m 38s
windows-latest-cmake (vulkan, -DSD_VULKAN=ON -DSD_BUILD_SHARED_LIBS=ON)	7m 47s
windows-latest-cmake-hip	3m 18s
windows-latest-cmake (avx2, -DGGML_NATIVE=OFF -DGGML_AVX2=ON -DSD_BUILD_SHARED_LIBS=ON)	3m 59s
windows-latest-cmake (noavx, -DGGML_NATIVE=OFF -DGGML_AVX=OFF -DGGML_AVX2=OFF -DGGML_FMA=OFF -DSD...	3m 40s
ubuntu-latest-rocm (7.2.1, gfx908;gfx90a;gfx942;gfx1030;gfx1031;gfx1032;gfx1100;gfx1101;gfx1102;g...	5m 4s
Build and push container images	0s
release	0s
	3h 10m 0s

Provide feedback