xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-05-17 21:46:52 +08:00
Code Issues Packages Projects Releases Wiki Activity

address ci issue

Browse Source 
Signed-off-by: bk-201 <joy25810@foxmail.com>
This commit is contained in:
bk-201 2025-12-10 15:25:23 +00:00
parent 5ff0c6fb73
commit d3c2f3dbe3
2 changed files with 7 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
vllm/lora/model_manager.py
View File

@ -65,8 +65,8 @@ class LoRAModelManager:
max_num_batched_tokens: int, max_num_batched_tokens: int,
vocab_size: int, vocab_size: int,
lora_config: LoRAConfig, lora_config: LoRAConfig,
vllm_config: VllmConfig,
device: torch.device, device: torch.device,
vllm_config: VllmConfig | None = None,
): ):
"""Create a LoRAModelManager and adapter for a given model. """Create a LoRAModelManager and adapter for a given model.
@ -114,7 +114,7 @@ class LoRAModelManager:
self.model.lora_manager = self self.model.lora_manager = self
def _init_multimodal_config(self, vllm_config: VllmConfig): def _init_multimodal_config(self, vllm_config: VllmConfig | None = None):
# Used to indicate whether the model is a multimodal model # Used to indicate whether the model is a multimodal model
self.supports_mm: bool = ( self.supports_mm: bool = (
supports_multimodal(self.model) supports_multimodal(self.model)
@ -125,7 +125,7 @@ class LoRAModelManager:
self.supports_mm_lora = False self.supports_mm_lora = False
if self.supports_mm: if self.supports_mm and vllm_config is not None:
model_config: ModelConfig = vllm_config.model_config model_config: ModelConfig = vllm_config.model_config
self.mm_mapping: MultiModelKeys = self.model.get_mm_mapping() self.mm_mapping: MultiModelKeys = self.model.get_mm_mapping()
if self.lora_config.enable_mm_lora: if self.lora_config.enable_mm_lora:
@ -708,8 +708,8 @@ class LRUCacheLoRAModelManager(LoRAModelManager):
max_num_batched_tokens: int, max_num_batched_tokens: int,
vocab_size: int, vocab_size: int,
lora_config: LoRAConfig, lora_config: LoRAConfig,
vllm_config: VllmConfig,
device: torch.device, device: torch.device,
vllm_config: VllmConfig | None = None,
): ):
super().__init__( super().__init__(
model, model,
@ -717,8 +717,8 @@ class LRUCacheLoRAModelManager(LoRAModelManager):
max_num_batched_tokens, max_num_batched_tokens,
vocab_size, vocab_size,
lora_config, lora_config,
vllm_config,
device, device,
vllm_config,
) )
self._registered_adapters: LoRALRUCache = LoRALRUCache( self._registered_adapters: LoRALRUCache = LoRALRUCache(
self.capacity, self.deactivate_adapter self.capacity, self.deactivate_adapter

4
vllm/lora/worker_manager.py
View File

@ -69,7 +69,7 @@ class WorkerLoRAManager:
def create_lora_manager( def create_lora_manager(
self, self,
model: torch.nn.Module, model: torch.nn.Module,
vllm_config: VllmConfig, vllm_config: VllmConfig | None = None,
) -> Any: ) -> Any:
lora_manager = create_lora_manager( lora_manager = create_lora_manager(
model, model,
@ -212,7 +212,7 @@ class LRUCacheWorkerLoRAManager(WorkerLoRAManager):
def create_lora_manager( def create_lora_manager(
self, self,
model: torch.nn.Module, model: torch.nn.Module,
vllm_config: VllmConfig, vllm_config: VllmConfig | None = None,
) -> Any: ) -> Any:
lora_manager = create_lora_manager( lora_manager = create_lora_manager(
model, model,