Primary and secondary liver cancer are the third cause of death in the world, and as the incidence is increasing, liver cancer represents a global health burden. Current treatment strategies are insufficient to permanently cure patients from this devastating disease, and therefore other approaches are under investigation. The importance of cancer-associated fibroblasts (CAFs) in the tumour microenvironment is evident, and many pre-clinical studies have shown increased tumour aggressiveness in the presence of CAFs. However, it remains unclear how hepatic stellate cells are triggered by the tumour to become CAFs and how the recently described CAF subtypes originate and orchestrate pro-tumoural effects. Specialized in vitro systems will be needed to address these questions. In this review, we present the currently used in vitro models to study CAFs in primary and secondary liver cancer and highlight the trend from using oversimplified 2D culture systems to more complex 3D models. Relatively few studies report on the impact of cancer (sub)types on CAFs and the tumour microenvironment, and most studies investigated the impact of secreted factors due to the nature of the models.