pull

source instead of exec in run-readme-pr-macos.yml #3913

Job	Run time
gather-models-cpu	11s
test-tinystories-executorch (16-core-ubuntu)	20m 7s
test-tinystories-executorch (macos-14-xlarge)	15m 10s
compile-gguf (macos-14)	7m 26s
gather-models-gpu	10s
runner-aoti (16-core-ubuntu)	17m 54s
runner-et (16-core-ubuntu)	8m 22s
runner-aoti (macos-14-xlarge)	3m 45s
runner-et (macos-14-xlarge)	7m 52s
test-mps (macos-m1-stable)	1m 40s
test-gguf-util (macos-14)	6m 7s
test-mps-dtype (macos-m1-stable)	2m 12s
test-torchao-experimental (macos-14-xlarge)	10m 47s
test-torchao-experimental-mps (macos-m1-stable)	1m 58s
torchchat-command-load-test (macos-14)	2m 36s
test-build-runner-et-android / linux-job	12m 13s
test-cpu-aoti (x86_64, stories15M)	22m 45s
test-cpu-compile (x86_64, stories15M)	17m 21s
test-cpu-aoti (aarch64, stories15M)	12m 6s
test-cpu-compile (aarch64, stories15M)	14m 11s
test-cpu-eval-sanity-check (x86_64, stories15M)	7m 51s
test-cpu-eval-sanity-check-float16 (x86_64, stories15M)	27m 3s
test-cpu-eval-sanity-check (aarch64, stories15M)	13m 7s
test-cpu-eval-sanity-check-float32 (x86_64, stories15M)	37m 17s
test-cpu-eval-sanity-check-float16 (aarch64, stories15M)	8m 10s
test-cpu-eval-sanity-check-float32 (aarch64, stories15M)	10m 35s
test-gpu-aoti-float16 (cuda, stories15M) / linux-job	14m 8s
test-gpu-aoti-bfloat16 (cuda, stories15M) / linux-job	15m 2s
test-gpu-aoti-float32 (cuda, stories15M) / linux-job	14m 11s
test-gpu-compile (cuda, stories15M) / linux-job	25m 32s
test-gpu-eval-sanity-check (cuda, stories15M) / linux-job	13m 54s
	6h 11m 43s

Provide feedback