FreedomIntelligence/ShizhenGPT-32B-VL
FreedomIntelligence/ShizhenGPT-32B-VL is a 32 billion parameter multimodal large language model developed by FreedomIntelligence, specifically designed for Traditional Chinese Medicine (TCM). This variant focuses on text and image understanding, derived from the ShizhenGPT-32B-Omni architecture. It provides strong expertise in TCM, supporting diagnostic capabilities through visual analysis and textual interaction. The model is suitable for applications requiring specialized TCM knowledge combined with visual input processing.
Loading preview...
ShizhenGPT-32B-VL: Multimodal LLM for Traditional Chinese Medicine
ShizhenGPT-32B-VL is a 32 billion parameter multimodal large language model developed by FreedomIntelligence, specializing in Traditional Chinese Medicine (TCM). It is a variant of the broader ShizhenGPT-32B-Omni model, specifically configured to handle text and image understanding tasks. This model is distinct as the first multimodal LLM tailored for TCM, integrating deep domain expertise with visual processing capabilities.
Key Capabilities
- TCM Expertise: Possesses strong knowledge in Traditional Chinese Medicine.
- Multimodal Understanding: Supports both textual and image inputs, enabling visual diagnostic assistance.
- Specialized Diagnostics: Facilitates TCM diagnostic processes, particularly those involving visual examination (望).
- Qwen2.5-VL Alignment: Its architecture aligns with Qwen2.5-VL, allowing for easier adaptation and deployment with tools like vllm or Sglang.
Good For
- Applications requiring specialized TCM knowledge.
- Tasks involving the interpretation of medical images within a TCM context.
- Developing AI assistants for TCM practitioners or educational tools.
- Research into multimodal AI for traditional medicine systems.
For broader multimodal needs, including the 'Four Diagnostics' (望闻问切), other ShizhenGPT-Omni variants are available or forthcoming. This VL version is recommended when use cases are primarily focused on text and vision.