ColossalAI
ColossalAI copied to clipboard
[DOC]: Hope to get your answer!
π The doc issue
two options:
- 8*24G A40900
- 8*48G A6000 .
I have LLM 11B, RewardModel 1B.
which option can run PPO model?
Bot detected the issue body's language is not English, translate it automatically. π―ππ»π§βπ€βπ§π«π§πΏβπ€βπ§π»π©πΎβπ€βπ¨πΏπ¬πΏ
Title: [DOC]: Hope to get your answer!
8*24G 4090 sorry!
Larger memory is more possible. You can try memory optimizations like zero, gemini, offload, etc. if encounter OOM.
We have updated a lot. This issue was closed due to inactivity. Thanks.