[Frontend] Require flag for loading text and image embeds (#27204)

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Co-authored-by: DarkLight1337 <tlleungac@connect.ust.hk>
This commit is contained in:
Russell Bryant
2025-10-22 15:52:02 +00:00
committed by GitHub
co-authored by DarkLight1337
parent db6f28d898
commit 58fab50d82
25 changed files with 203 additions and 64 deletions
+10
View File
@@ -1308,6 +1308,16 @@ class BaseMultiModalProcessor(ABC, Generic[_I]):
[`_get_hf_mm_data`][vllm.multimodal.processing.BaseMultiModalProcessor._get_hf_mm_data].
"""
mm_items = self.data_parser.parse_mm_data(mm_data)
mm_config = self.info.ctx.model_config.get_multimodal_config()
if not mm_config.enable_mm_embeds:
for modality, items in mm_items.items():
if isinstance(items, (EmbeddingItems, DictEmbeddingItems)):
raise ValueError(
f"You must set `--enable-mm-embeds` to input "
f"`{modality}_embeds`"
)
for modality, items in mm_items.items():
self.validate_num_items(modality, len(items))