Janus Pro 7b 4bit

🚀 Janus-Pro

Janus-Pro is a novel autoregressive framework that unifies multimodal understanding and generation, offering high flexibility and effectiveness.

🚀 Quick Start

Please refer to Github Repository

✨ Features

• Janus-Pro is a novel autoregressive framework that unifies multimodal understanding and generation.
• It decouples visual encoding into separate pathways while using a single, unified transformer architecture, alleviating the conflict between the visual encoder’s roles in understanding and generation and enhancing flexibility.
• It surpasses previous unified models and matches or exceeds the performance of task - specific models.
• It is a unified understanding and generation MLLM, constructed based on the DeepSeek - LLM - 1.5b - base/DeepSeek - LLM - 7b - base.

Github Repository

📚 Documentation

Model Summary

Janus-Pro is a unified understanding and generation MLLM, which decouples visual encoding for multimodal understanding and generation. Janus-Pro is constructed based on the DeepSeek-LLM-1.5b-base/DeepSeek-LLM-7b-base.

For multimodal understanding, it uses the SigLIP-L as the vision encoder, which supports 384 x 384 image input. For image generation, Janus-Pro uses the tokenizer from here with a downsample rate of 16.

📄 License

This code repository is licensed under the MIT License. The use of Janus-Pro models is subject to DeepSeek Model License.

🔧 Technical Details

Janus-Pro addresses the limitations of previous approaches by decoupling visual encoding into separate pathways, while still utilizing a single, unified transformer architecture for processing. The decoupling not only alleviates the conflict between the visual encoder’s roles in understanding and generation, but also enhances the framework’s flexibility.

📚 Citation

@misc{chen2025januspro,
      title={Janus-Pro: Unified Multimodal Understanding and Generation with Data and Model Scaling}, 
      author={Xiaokang Chen and Zhiyu Wu and Xingchao Liu and Zizheng Pan and Wen Liu and Zhenda Xie and Xingkai Yu and Chong Ruan},
      year={2025},
}

📞 Contact

If you have any questions, please raise an issue or contact us at service@deepseek.com.