https://github.com/bytedance/f-16

F-16 is a powerful video large language model (LLM) that perceives high-frame-rate videos, which is developed by the Department of Electronic Engineering at Tsinghua University and ByteDance.

https://github.com/bytedance/f-16

Science Score: 36.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • CITATION.cff file
  • codemeta.json file
    Found codemeta.json file
  • .zenodo.json file
    Found .zenodo.json file
  • DOI references
  • Academic publication links
    Links to: arxiv.org
  • Academic email domains
  • Institutional organization owner
  • JOSS paper metadata
  • Scientific vocabulary similarity
    Low similarity (6.7%) to scientific vocabulary

Keywords

research
Last synced: 10 months ago · JSON representation

Repository

F-16 is a powerful video large language model (LLM) that perceives high-frame-rate videos, which is developed by the Department of Electronic Engineering at Tsinghua University and ByteDance.

Basic Info
  • Host: GitHub
  • Owner: bytedance
  • License: apache-2.0
  • Language: Python
  • Default Branch: main
  • Homepage:
  • Size: 70.3 KB
Statistics
  • Stars: 3
  • Watchers: 0
  • Forks: 1
  • Open Issues: 0
  • Releases: 0
Topics
research
Created about 1 year ago · Last pushed about 1 year ago
Metadata Files
Readme License

README.md

Improving LLM Video Understanding with 16 Frames Per Second

🚀🚀 Welcome to the repo of F-16!

F-16 is a powerful video large language model (LLM) that perceives high-frame-rate videos, which is developed by the Department of Electronic Engineering at Tsinghua University and ByteDance.

🔥 News

  • 2025-07-03: We release the final checkpoint of F-16.
  • 2025-06-18: We release the code of F-16.

⚡️ Future Plans

  • ~~Release the code.~~
  • ~~Release final F-16.~~

🌈 How to Use

How to train a model

  1. Prepare the dataset following scripts/example_sft.json.
  2. Download LLaVA-OneVision Model from huggingface.
  3. Modify the parameters in scripts/train_sft.sh.
  4. Run bash scripts/train_sft.sh.

How to evaluate a checkpoint

  1. Prepare the dataset following scripts/example_sft.json.
  2. Modify the parameters in scripts/eval.sh.
  3. Run bash scripts/eval.sh.

👀 Team

Team Tsinghua: Yixuan Li, Changli Tang, Jimin Zhuang, Yudong Yang, Guangzhi Sun, Chao Zhang

Team ByteDance: Wei Li, Zejun Ma

✨ Citation

If you find F-16 useful, please cite the paper:

@inproceedings{li2025improving, title={Improving LLM Video Understanding with 16 Frames Per Second}, author={Li, Yixuan and Tang, Changli and Zhuang, Jimin and Yang, Yudong and Sun, Guangzhi and Li, Wei and Ma, Zejun and Zhang, Chao}, booktitle={Proc. ICML}, year={2025}, address={Vancouver} }

Owner

  • Name: Bytedance Inc.
  • Login: bytedance
  • Kind: organization
  • Location: Singapore

GitHub Events

Total
  • Watch event: 5
  • Push event: 1
  • Fork event: 2
Last Year
  • Watch event: 5
  • Push event: 1
  • Fork event: 2

Dependencies

requirements.txt pypi
  • transformers ==4.39.2