[Feature] Support unified paging in multi-lora serving
Checklist
- [x] 1. If the issue you raised is not a feature but a question, please raise a discussion at https://github.com/sgl-project/sglang/discussions/new/choose Otherwise, it will be closed.
- [x] 2. Please use English, otherwise it will be closed.
Motivation
Currently, SGL doesn't support the unified paging feature proposed by S-LoRA. However, this feature is important for memory management in multi-LoRA serving.
Related resources
No response
We will try to implement it soon.
Link: https://github.com/sgl-project/sglang/issues/2929
Sure! Thank you for your help~ cc @Fridge003
@Sunt-ing Thanks! Looking forward to your contribution!
This issue has been automatically closed due to inactivity. Please feel free to reopen it if needed.
This issue has been automatically closed due to inactivity. Please feel free to reopen it if needed.
This issue has been automatically closed due to inactivity. Please feel free to reopen it if needed.
This issue has been automatically closed due to inactivity. Please feel free to reopen it if needed.