Despite the success of large language models (LLMs), their potential in **Traditional Chinese Medicine (TCM)** remains underexplored due to scarce high-quality data and the multimodal nature of TCM diagnostics, which involve looking, listening, smelling, and pulse-taking. These modalities challenge conventional LLMs. To address these challenges, we present **ShizhenGPT**, the first multimodal LLM tailored for TCM. To overcome data scarcity, we curate the largest TCM dataset to date, comprising 100GB+ of text and 200GB+ of multimodal data, including 1.2M images, 200 hours of audio, and physiological signals. ShizhenGPT is pretrained and instruction-tuned to achieve deep TCM knowledge and multimodal reasoning. For evaluation, we collect recent national TCM qualification exams and build a visual benchmark for Medicinal Recognition and Visual Diagnosis. Experiments demonstrate that ShizhenGPT outperforms comparable-scale LLMs and competes with larger proprietary models. Moreover, it leads in TCM visual understanding among existing multimodal LLMs and shows promising results across modalities such as sound, pulse, electronic-nose signals, and vision, paving the way toward more holistic multimodal perception in TCM. We hope that the datasets, models, and methods of this work will inspire further exploration in this field. Code, models, and released data are available at https://github.com/FreedomIntelligence/ShizhenGPT.