[VLM] Remove BaseProcessingInfo.get_mm_max_tokens_per_item (#16408)

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
This commit is contained in:
Cyrus Leung
2025-04-10 09:06:58 -07:00
committed by GitHub
parent 7678fcd5b6
commit 83b824c8b4
39 changed files with 104 additions and 677 deletions
@@ -61,22 +61,6 @@ class LlavaNextVideoProcessingInfo(BaseProcessingInfo):
def get_supported_mm_limits(self) -> Mapping[str, Optional[int]]:
return {"video": 1}
def get_mm_max_tokens_per_item(
self,
seq_len: int,
mm_counts: Mapping[str, int],
) -> Mapping[str, int]:
target_width, target_height = self.get_image_size_with_most_features()
max_video_tokens = self.get_num_video_tokens(
image_width=target_width,
image_height=target_height,
num_frames=self.get_num_frames_with_most_features(
seq_len, mm_counts),
)
return {"video": max_video_tokens}
def get_image_size_with_most_features(self) -> ImageSize:
vision_encoder_info = self.get_vision_encoder_info()
width = height = vision_encoder_info.get_image_size()