pull

Enable sdpa backends for server export in export.py #3871

Job	Run time
gather-models-cpu	11s
gather-models-gpu	12s
compile-gguf (macos-14)	7m 35s
runner-aoti (16-core-ubuntu)	6m 40s
runner-et (16-core-ubuntu)	9m 1s
runner-aoti (macos-14-xlarge)	3m 32s
runner-et (macos-14-xlarge)	7m 26s
test-gguf-util (macos-14)	5m 22s
test-mps (macos-m1-stable)	1m 25s
test-mps-dtype (macos-m1-stable)	1m 53s
test-tinystories-executorch (16-core-ubuntu)	19m 8s
test-torchao-experimental (macos-14-xlarge)	10m 55s
test-torchao-experimental-mps (macos-m1-stable)	53s
test-tinystories-executorch (macos-14-xlarge)	15m 2s
torchchat-command-load-test (macos-14)	2m 12s
test-build-runner-et-android / linux-job	12m 25s
test-cpu-aoti (x86_64, stories15M)	22m 57s
test-cpu-eval-sanity-check (x86_64, stories15M)	7m 16s
test-cpu-eval-sanity-check-float16 (x86_64, stories15M)	27m 37s
test-cpu-aoti (aarch64, stories15M)	9m 34s
test-cpu-eval-sanity-check (aarch64, stories15M)	10m 16s
test-cpu-compile (x86_64, stories15M)	17m 40s
test-cpu-eval-sanity-check-float16 (aarch64, stories15M)	8m 20s
test-cpu-compile (aarch64, stories15M)	10m 36s
test-cpu-eval-sanity-check-float32 (x86_64, stories15M)	37m 27s
test-cpu-eval-sanity-check-float32 (aarch64, stories15M)	11m 3s
test-gpu-compile (cuda, stories15M) / linux-job	25m 5s
test-gpu-aoti-bfloat16 (cuda, stories15M) / linux-job	14m 56s
test-gpu-aoti-float16 (cuda, stories15M) / linux-job	14m 25s
test-gpu-aoti-float32 (cuda, stories15M) / linux-job	14m 0s
test-gpu-eval-sanity-check (cuda, stories15M) / linux-job	13m 46s
	5h 48m 50s

Provide feedback