Chinese LLaVA Cllama2

🚀 Chinese LLaVA

An open-source and commercially available Chinese-English bilingual vision-language assistant, Chinese-LLaVA, along with the Chinese-English vision SFT dataset, Chinese-LLaVA-Vision-Instructions. It's an open-source and commercially available dialogue model supporting Chinese-English vision-text multimodal conversations.

🚀 Quick Start

✨ Features

• Open source and commercially available.
• Supports Chinese-English bilingual vision-language interaction.
• Comes with a dedicated vision SFT dataset.

📦 Installation

# clone the repository
git clone https://github.com/LinkSoul-AI/Chinese-LLaVA
cd Chinese-LLaVA

# install package
conda create -n Cllava python=3.10 -y
conda activate Cllava
pip install --upgrade pip
pip install -e .

💻 Usage Examples

Basic Usage

python infer.py \
    --model-name PATH/TO/THE/CHINESE_LLAVA_MODEL \
    --llm-type "Chinese_llama2" or "baichuan" \
    --image-file PATH/TO/THE/INPUT/IMAGE \
    --query QUERY/PROMPT

🔍 Details

Base Demo

Online Demo

Talk is cheap, Show you the Demo.

• Demo Address / HuggingFace Spaces

Resources Download

• Models:

  • Chinese-LLaVA-Chinese-Llama-2-7B
  • Chinese-LLaVA-Baichuan-7B

• Baidu Netdisk Download:

  • Chinese-LLaVA-Chinese-Llama-2-7B
  • Chinese-LLaVA-Baichuan-7B

• Language Models:

  • Chinese-Llama-2-7b
  • Baichuan-7B

• Dataset: Chinese-LLaVA-Vision-Instructions

📄 License

Apache-2.0 license

📚 Related Projects

• LLaVA
• Chinese-Llama-2-7B
• baichuan-inc/Baichuan-7B

💡 Future Plans

• How to train the model.
• int4 quantization.
• Docker deployment.

💬 WeChat Group

Welcome to join the WeChat Group