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

[Misc] Clean up type annotation for SupportsMultiModal (#14794)

Browse Source 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
This commit is contained in:
Cyrus Leung
2025-03-14 15:59:56 +08:00
committed by GitHub
parent 09269b3127
commit 601bd3268e
27 changed files with 121 additions and 141 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
vllm/model_executor/models/llava.py
View File

@@ -38,7 +38,7 @@ from vllm.sequence import IntermediateTensors
from vllm.utils import JSONTree, flatten_2d_lists, json_map_leaves
from .clip import CLIPVisionModel
from .interfaces import SupportsMultiModal, SupportsPP
from .interfaces import MultiModalEmbeddings, SupportsMultiModal, SupportsPP
from .pixtral import (PixtralHFVisionModel,
get_pixtral_hf_image_feature_grid_size)
from .siglip import SiglipVisionModel
@@ -778,7 +778,8 @@ class LlavaForConditionalGeneration(nn.Module, SupportsMultiModal, SupportsPP):
return embeds_in_batch
def get_multimodal_embeddings(self, **kwargs) -> Optional[NestedTensors]:
def get_multimodal_embeddings(
self, **kwargs: object) -> Optional[MultiModalEmbeddings]:
image_input = self._parse_and_validate_image_input(**kwargs)
if image_input is None:
return None
@@ -800,7 +801,7 @@ class LlavaForConditionalGeneration(nn.Module, SupportsMultiModal, SupportsPP):
def get_input_embeddings(
self,
input_ids: torch.Tensor,
multimodal_embeddings: Optional[NestedTensors] = None,
multimodal_embeddings: Optional[MultiModalEmbeddings] = None,
) -> torch.Tensor:
inputs_embeds = self.language_model.get_input_embeddings(input_ids)
if multimodal_embeddings is not None: