In order to make reliable predictions of thermal cracking risks in young concrete, modelling of the creep behaviour is important. The linear logarithmic creep model (LLM) has only been used to describe the creep behaviour of moisture-sealed concrete samples (basic creep). The aim of the present research is to check whether LLM could be also applied to drying creep. The results indicate that the LLM delivers good results for sealed but not for drying samples. Thus, refinement of the existing model (e.g., tri-linear instead of bi-linear in logarithmic time scale), or a new model is needed to account for the increase of creep due to on-going drying.