7 Commits

Include renames

Author	SHA1	Message	Date
tingxueronghua	9ed87aa688	support fp8 model loading and cpu offload strategy. Thus the workflow could run in 24GB GPU.	2025-04-30 14:54:42 +08:00
Quank	8ac97335cf	Update step1xeditnode.py	2025-04-28 19:34:19 +08:00
Quank	4dd03e8c13	Fix errors	2025-04-28 18:12:42 +08:00
Quank	6c124a0252	Update step1xeditnode.py	2025-04-28 11:16:23 +08:00
Quank	a68a478813	Update step1xeditnode.py	2025-04-28 11:12:50 +08:00
Quank	e663094f8e	Fix import issues	2025-04-28 09:02:43 +08:00
quank123wip	c4ab9716c2	init repo	2025-04-26 23:20:54 +08:00