File size: 4,145 Bytes
60d8849 770d876 60d8849 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 | ---
license: apache-2.0
datasets:
- FreedomIntelligence/TCM-Pretrain-Data-ShizhenGPT
- FreedomIntelligence/TCM-Instruction-Tuning-ShizhenGPT
language:
- zh
base_model:
- Qwen/Qwen2.5-32B
pipeline_tag: text-generation
tags:
- Traditional Chinese Medicin
- Multimodal LLM
- multimodal
---
<div align="center">
<h1>
ShizhenGPT-32B-LLM
</h1>
</div>
<div align="center">
<a href="https://github.com/FreedomIntelligence/ShizhenGPT" target="_blank">GitHub</a> | <a href="https://arxiv.org/abs/2508.14706" target="_blank">Paper</a>
</div>
**ShizhenGPT** is the first multimodal LLM for Traditional Chinese Medicine (TCM).
It not only possesses strong expertise in TCM, but also supports TCM multimodal diagnostic capabilities, which involve looking (望), listening/smelling (闻), questioning (问), and pulse-taking (切).
👉 More details on GitHub: [ShizhenGPT](https://github.com/FreedomIntelligence/ShizhenGPT)
# <span>Model Info</span>
> **ShizhenGPT-32B-LLM** is an LLM-only variant derived from ShizhenGPT-32B-Omni. If your use case requires only text-based capabilities, this version is recommended. Otherwise, please choose the appropriate multimodal version below:
| | Parameters | Supported Modalities | Link |
| ---------------------- | ---------- | ----------------------------- | --------------------------------------------------------------------- |
| **ShizhenGPT-7B-LLM** | 7B | Text | [HF Link](https://huggingface.co/FreedomIntelligence/ShizhenGPT-7B-LLM) |
| **ShizhenGPT-7B-VL** | 7B | Text, Image Understanding | [HF Link](https://huggingface.co/FreedomIntelligence/ShizhenGPT-7B-VL) |
| **ShizhenGPT-7B-Omni** | 7B | Text, Four Diagnostics (望闻问切) | [HF Link](https://huggingface.co/FreedomIntelligence/ShizhenGPT-7B-Omni) |
| **ShizhenGPT-32B-LLM** | 32B | Text | [HF Link](https://huggingface.co/FreedomIntelligence/ShizhenGPT-32B-LLM) |
| **ShizhenGPT-32B-VL** | 32B | Text, Image Understanding | [HF Link](https://huggingface.co/FreedomIntelligence/ShizhenGPT-32B-VL) |
| **ShizhenGPT-32B-Omni** | 32B | Text, Four Diagnostics (望闻问切) | Available soon |
*Note: The LLM and VL models are parameter-split variants of ShizhenGPT-32B-Omni. Since their architectures align with Qwen2.5 and Qwen2.5-VL, they are easier to adapt to different environments. In contrast, ShizhenGPT-7B-Omni requires `transformers==4.51.0`.*
# <span>Usage</span>
You can use ShizhenGPT-32B-LLM in the same way as [Qwen2.5-32B-Instruct](https://huggingface.co/Qwen/Qwen2.5-32B-Instruct). You can deploy it with tools like [vllm](https://github.com/vllm-project/vllm) or [Sglang](https://github.com/sgl-project/sglang), or perform direct inference:
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
model = AutoModelForCausalLM.from_pretrained("FreedomIntelligence/ShizhenGPT-32B-LLM",torch_dtype="auto",device_map="auto")
tokenizer = AutoTokenizer.from_pretrained("FreedomIntelligence/ShizhenGPT-32B-LLM")
input_text = "为什么我总是手脚冰凉，是阳虚吗？"
messages = [{"role": "user", "content": input_text}]
inputs = tokenizer(tokenizer.apply_chat_template(messages, tokenize=False,add_generation_prompt=True
), return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=2048)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```
# <span>📖 Citation</span>
```
@misc{chen2025shizhengptmultimodalllmstraditional,
title={ShizhenGPT: Towards Multimodal LLMs for Traditional Chinese Medicine},
author={Junying Chen and Zhenyang Cai and Zhiheng Liu and Yunjin Yang and Rongsheng Wang and Qingying Xiao and Xiangyi Feng and Zhan Su and Jing Guo and Xiang Wan and Guangjun Yu and Haizhou Li and Benyou Wang},
year={2025},
eprint={2508.14706},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2508.14706},
}
``` |