fix(kokoro): accept published GGUF tensor names + emit F16 weights (#9588) #206

Job	Run time
android-arm64	5m 23s
macOS-latest-x64	4m 58s
macOS-latest-arm64	2m 19s
ggml-ci-x64-cpu-low-perf	1m 48s
macOS-latest-arm64-webgpu	16m 12s
ubuntu-24-vulkan (arm64, ubuntu-24.04-arm)	8m 17s
ubuntu-cpu-riscv64-native	1d 0h 0m 1s
ubuntu-latest-cuda	28m 47s
ubuntu-cpu (arm64, ubuntu-24.04-arm)	2m 10s
ggml-ci-arm64-cpu-high-perf-sve	4m 28s
ubuntu-24-webgpu-wasm	3m 40s
ggml-ci-arm64-cpu-low-perf	1m 54s
ubuntu-24-vulkan (x64, ubuntu-24.04)	13m 52s
windows-latest-hip	3m 8s
build-cmake-pkg / linux	9m 41s
ggml-ci-x64-cpu-high-perf	2m 40s
ggml-ci-arm64-cpu-kleidiai-graviton4	1d 0h 0m 1s
windows-latest (vulkan-x64, x64, -DCMAKE_BUILD_TYPE=Release -DGGML_NATIVE=OFF -DLLAMA_BUILD_SERVE...	17m 9s
ubuntu-22-musa	16m 15s
ubuntu-cpu (s390x, ubuntu-24.04-s390x)	2m 53s
ubuntu-cpu (x64, ubuntu-22.04)	2m 31s
windows-latest (openblas-x64, x64, -G "Ninja Multi-Config" -D CMAKE_TOOLCHAIN_FILE=cmake/x64-wind...	13m 15s
windows-latest (llvm-arm64-opencl-adreno, arm64, -G "Ninja Multi-Config" -D CMAKE_TOOLCHAIN_FILE=...	7m 18s
ubuntu-cpu (ppc64le, ubuntu-24.04-ppc64le)	2m 53s
ggml-ci-arm64-cpu-kleidiai	3m 5s
windows-2022-cuda (12.4)	2h 39m 56s
ubuntu-latest-rpc	7m 14s
ubuntu-24-webgpu	9m 32s
ggml-ci-arm64-cpu-high-perf	4m 6s
windows-latest (cpu-x64 (static), x64, -G "Ninja Multi-Config" -D CMAKE_TOOLCHAIN_FILE=cmake/x64-...	7m 2s
ubuntu-22-hip	17m 16s
windows-latest (llvm-arm64, arm64, -G "Ninja Multi-Config" -D CMAKE_TOOLCHAIN_FILE=cmake/arm64-wi...	7m 30s
	2d 6h 27m 14s

Navigation Menu