Hugging Face's logo

moonshotai
/
Kimi-K2.5

Image-Text-to-Text
Transformers
Safetensors
kimi_k25
feature-extraction
compressed-tensors
conversational
custom_code
Eval Results
Model card Files
xet
Community
53

fix: type annotation bugs in media_utils, processor and config

#35
by ThanhNguyxn - opened
base: refs/heads/main
←
from: refs/pr/35
Discussion Files changed
+569
-569
ThanhNguyxn

Summary

  • Fix VideoSpec.media_type: use Literal['video'] correctly instead of assigning type to str
  • Fix VideoChunkInput TypedDict: use NotRequired for optional prompt field (Python 3.11+)
  • Fix preprocess_medias return type: tuple[list[dict], list[str]] instead of list[dict]
  • Fix KimiK25Config.init: proper Optional types for text_config and vision_config
  • Add missing type annotation for video_placeholder parameter
 fix: type annotation bugs in media_utils, processor and config66cbfac6
Ready to merge
This branch is ready to get merged automatically.

Sign up or log in to comment