chore: ⬆️ Update vllm-metal (darwin) to `v0.3.0.dev2026081508… #3392

Job	Run time
generate-matrix	10s
backend-jobs-singlearch-1 (linux/amd64, auto, -cpu-vllm, ubuntu-latest, ubuntu:24.04, true, vllm,... / backend-build	7m 38s
backend-jobs-singlearch-4 (intel, linux/amd64, auto, -gpu-intel-vllm, ubuntu-latest, intel/oneapi... / backend-build	33m 35s
backend-jobs-singlearch-1 (cublas, 12, 8, linux/amd64, auto, -gpu-nvidia-cuda-12-vllm, ubuntu-lat... / backend-build	21m 29s
backend-jobs-singlearch-3 (hipblas, linux/amd64, auto, -gpu-rocm-hipblas-vllm, ubuntu-latest, roc... / backend-build	30m 56s
backend-jobs-darwin (vllm, -metal-darwin-arm64-vllm, mps) / darwin-backend-build (1.25.x)	15m 14s
backend-jobs-singlearch-2 (cublas, 13, 0, linux/amd64, auto, -gpu-nvidia-cuda-13-vllm, ubuntu-lat... / backend-build	17m 47s
backend-jobs-singlearch-2 (l4t, 13, 0, linux/arm64, auto, -nvidia-l4t-cuda-13-arm64-vllm, ubuntu-... / backend-build	15m 53s
backend-jobs-multiarch	0s
backend-merge-jobs-multiarch	0s
backend-jobs-darwin (vllm, -metal-darwin-arm64-vllm, mps) / darwin-backend-publish	1m 21s
backend-merge-jobs-singlearch-2 (-nvidia-l4t-cuda-13-arm64-vllm, auto) / merge	41s
backend-merge-jobs-singlearch-2 (-gpu-nvidia-cuda-13-vllm, auto) / merge	34s
backend-merge-jobs-singlearch-1 (-gpu-nvidia-cuda-12-vllm, auto) / merge	52s
backend-merge-jobs-singlearch-1 (-cpu-vllm, auto) / merge	40s
backend-merge-jobs-singlearch-4 (-gpu-intel-vllm, auto) / merge	50s
backend-merge-jobs-singlearch-3 (-gpu-rocm-hipblas-vllm, auto) / merge	44s
	2h 28m 24s

Navigation Menu