Skip to content

Feat (gptq): optimizing CPU to GPU memory transfer #1733

Feat (gptq): optimizing CPU to GPU memory transfer

Feat (gptq): optimizing CPU to GPU memory transfer #1733

Job Run time
15m 18s
23m 57s
14m 28s
1h 7m 6s
28m 59s
1h 8m 4s
10m 45s
36m 4s
15m 8s
1h 6m 23s
24m 43s
1h 8m 47s
10m 37s
41m 10s
17m 37s
1h 12m 58s
28m 4s
1h 12m 30s
10m 36s
39m 35s
17m 45s
1h 4m 58s
25m 56s
1h 4m 33s
10m 42s
30m 34s
18m 20s
1h 6m 29s
25m 57s
1h 4m 41s
12m 7s
37m 1s
18m 8s
1h 5m 50s
27m 44s
1h 3m 16s
12m 0s
35m 18s
14m 40s
25m 41s
15m 38s
1h 7m 27s
27m 6s
1h 5m 41s
9m 38s
38m 24s
15m 26s
1h 6m 34s
24m 41s
1h 8m 5s
10m 16s
46m 54s
17m 23s
1h 9m 57s
27m 14s
1h 11m 8s
10m 30s
40m 31s
17m 55s
1h 4m 23s
25m 54s
1h 4m 12s
10m 19s
34m 2s
19m 16s
1h 6m 11s
26m 5s
1h 4m 4s
12m 2s
35m 19s
17m 56s
1h 4m 26s
26m 1s
1h 2m 29s
10m 59s
34m 38s
1d 22h 23m 13s