BigHugger
GH Repository · OpenGVLab

Ask-Anything

[CVPR2024 Highlight][VideoChatGPT] ChatGPT with video understanding! And many more supported LMs such as miniGPT4, StableLM, and MOSS.

stars
3,351
30-day movement
starts with the next reading
Related entries
60
Connections
1
agent-frameworkPythonlangchainlarge-language-modelschatgptgradiovideo-understandingcaptioning-videosstablelmchatlarge-modelvideovideo-question-answeringbig-modelfoundation-models

Ask-Anything is a Python project from OpenGVLab that combines ChatGPT-style chat with video understanding, described as VideoChatGPT. It also supports other large language models such as miniGPT4, StableLM, and MOSS, and was highlighted at CVPR2024.

Reach for it when you want to ask questions about videos or caption them using large language models through a chat interface.

Use it to

  • Chat with a model about video content
  • Generate captions for videos
  • Answer questions about videos
  • Swap in alternative LMs like StableLM or MOSS
  • Run demos locally via its Gradio interface

For Researchers and developers working on video-language models

Role
agent-framework
Language
Python
Licence
MIT
Forks
268
Open issues
70
Last push
2026-07-17
topicsvideo-understandingvideo-question-answeringlarge-language-modelschatgptcaptioning-videosgradio