Back to pull request #2630

CUDA-graph-compatible releasing and resuming KV cache and model weight memory #3532

Sign in to view logs

Run time

Learn about OS pricing on GitHub Actions

Job	Run time
unit-test-frontend	3m 47s
performance-test-1-gpu-part-1	9m 48s
accuracy-test-1-gpu	5m 21s
accuracy-test-2-gpu	6m 22s
performance-test-1-gpu-part-2	11m 25s
performance-test-2-gpu	9m 52s
unit-test-backend-1-gpu (0-6)	16m 12s
unit-test-backend-2-gpu	13m 16s
unit-test-backend-1-gpu (6-16)	12m 20s
unit-test-backend-1-gpu (16-23)	13m 19s
unit-test-backend-1-gpu (23-30)	13m 5s
unit-test-backend-1-gpu (30-100)	15m 20s
finish	1s
	2h 10m 8s