FreedomIntelligence/ShizhenGPT-7B-VL

ShizhenGPT-7B-VL: Multimodal LLM for Traditional Chinese Medicine

ShizhenGPT-7B-VL is a specialized 7 billion parameter multimodal large language model developed by FreedomIntelligence, focusing on Traditional Chinese Medicine (TCM). It is derived from the broader ShizhenGPT-7B-Omni model but is specifically tailored for applications requiring text and image understanding. This model is the first of its kind to integrate strong TCM expertise with multimodal diagnostic capabilities, such as interpreting visual information like tongue coatings.

Key Capabilities

• TCM Expertise: Possesses deep knowledge in Traditional Chinese Medicine.
• Multimodal Understanding: Capable of processing both textual prompts and image inputs.
• Vision-Language Integration: Designed for tasks that involve analyzing images in a TCM context, such as visual diagnostics.
• Qwen2.5-VL Alignment: Its architecture aligns with Qwen2.5-VL, facilitating easier adaptation and deployment in various environments.

Good For

• TCM-specific AI applications: Ideal for developing tools that require understanding and generating content related to Traditional Chinese Medicine.
• Image-based TCM diagnostics: Suitable for tasks like analyzing medical images (e.g., tongue or facial diagnosis) from a TCM perspective.
• Research and development in TCM AI: Provides a foundation for further exploration into AI applications within the field of Traditional Chinese Medicine.
• Text-only or vision-only tasks: Recommended when your application primarily involves text or image processing within the TCM domain, offering a more streamlined solution compared to the full Omni version.