biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[V0 deprecation] Remove long context LoRA (#21169)

Browse Source 
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
This commit is contained in:
Jee Jee Li
2025-07-19 17:15:41 +08:00
committed by GitHub
parent cf8cc32674
commit 1eaff27815
13 changed files with 35 additions and 301 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
vllm/engine/arg_utils.py
View File

@@ -358,8 +358,6 @@ class EngineArgs:
max_cpu_loras: Optional[int] = LoRAConfig.max_cpu_loras
lora_dtype: Optional[Union[str, torch.dtype]] = LoRAConfig.lora_dtype
lora_extra_vocab_size: int = LoRAConfig.lora_extra_vocab_size
long_lora_scaling_factors: Optional[tuple[float, ...]] = \
LoRAConfig.long_lora_scaling_factors
# PromptAdapter fields
enable_prompt_adapter: bool = False
max_prompt_adapters: int = PromptAdapterConfig.max_prompt_adapters
@@ -723,8 +721,6 @@ class EngineArgs:
"--lora-dtype",
**lora_kwargs["lora_dtype"],
)
lora_group.add_argument("--long-lora-scaling-factors",
**lora_kwargs["long_lora_scaling_factors"])
lora_group.add_argument("--max-cpu-loras",
**lora_kwargs["max_cpu_loras"])
lora_group.add_argument("--fully-sharded-loras",
@@ -1245,7 +1241,6 @@ class EngineArgs:
default_mm_loras=self.default_mm_loras,
fully_sharded_loras=self.fully_sharded_loras,
lora_extra_vocab_size=self.lora_extra_vocab_size,
long_lora_scaling_factors=self.long_lora_scaling_factors,
lora_dtype=self.lora_dtype,
max_cpu_loras=self.max_cpu_loras if self.max_cpu_loras
and self.max_cpu_loras > 0 else None) if self.enable_lora else None