support vllm

This commit is contained in:
hiyouga
2024-03-07 20:26:31 +08:00
parent f74f804a71
commit d07ad5cc1c
32 changed files with 752 additions and 316 deletions

View File

@@ -1,3 +1,4 @@
Usage:
- `merge.sh` -> `quantize.sh`
- `merge.sh`: merge the lora weights
- `quantize.sh`: quantize the model with AutoGPTQ (must after merge.sh, optional)