Views
No views yet
git clone https://github.com/FreedomIntelligence/LongLLaVA.gitpip install -r requirements.txtpython cli.py --model_dir path-to-longllava1query = 'What does the picture show?'
2image_paths = ['image_path1'] # image or video path
3
4from cli import Chatbot
5bot = Chatbot(path-to-longllava)
6output = bot.chat(query, image_paths)
7print(output) # Prints the output of the model@misc{wang2024longllavascalingmultimodalllms,
title={LongLLaVA: Scaling Multi-modal LLMs to 1000 Images Efficiently via Hybrid Architecture},
author={Xidong Wang and Dingjie Song and Shunian Chen and Chen Zhang and Benyou Wang},
year={2024},
eprint={2409.02889},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2409.02889},
}