Support limiting max loaded loras in CPU. (#8650)

This commit is contained in:
Lifu Huang
2025-08-03 00:02:23 -07:00
committed by GitHub
parent a437aa9987
commit 8675bdf246
8 changed files with 163 additions and 58 deletions

View File

@@ -514,6 +514,7 @@ class SRTRunner:
max_lora_rank: Optional[int] = None,
lora_target_modules: Optional[List[str]] = None,
enable_lora: Optional[bool] = None,
max_loaded_loras: Optional[int] = None,
):
self.model_type = model_type
self.is_generation = model_type == "generation"
@@ -556,6 +557,7 @@ class SRTRunner:
max_lora_rank=max_lora_rank,
lora_target_modules=lora_target_modules,
enable_lora=enable_lora,
max_loaded_loras=max_loaded_loras,
**spec_kwargs,
)