With the rapid development of artificial intelligence (AI) within medicine, the emergence of large language models (LLMs) has gradually reached the forefront of clinical research. In oncology, by mining the underlying connection between a text or image input and the desired output, LLMs demonstrate great potential for managing tumours. In this review, we provide a brief description of the development of LLMs, followed by model construction strategies and general medical functions. We then elaborate on the role of LLMs in cancer screening and diagnosis, metastasis identification, tumour staging, treatment recommendation, and documentation processing tasks by decoding various types of clinical data. Moreover, the current barriers faced by LLMs, such as hallucinations, ethical problems, limited application, and so on, are outlined along with corresponding solutions, where the further purpose is to inspire improvement and innovation in this field with respect to harnessing LLMs for advancing precision oncology.Copyright © 2025 The Author(s). Published by Elsevier B.V. All rights reserved.