Background and purpose: The influence of bearing on short-term revision in press-fit total hip arthroplasty (THA) remains under-reported. The aim of this study was to describe 2-year cup revision rates of ceramic-on-ceramic (CoC) and ceramic-on-polyethylene (CoPE). Patients and methods: Primary press-fit THAs with one of the three most used cups available with both CoC or CoPE bearing recorded in the Dutch Arthroplasty Register (LROI) were included (2007–2019). Primary outcome was 2-year cup revision for all reasons. Secondary outcomes were: reasons for revision, incidence of different revision procedures and use of both bearings over time. Results: 2-year Kaplan-Meier cup revision rate in 33,454 THAs (12,535 CoC; 20,919 CoPE) showed a higher rate in CoC (0.67% [95% CI, 0.54–0.81]) compared to CoPE (0.44% [95% CI, 0.34–0.54]) ( p = 0.004). Correction for confounders (age, gender, cup type, head size) resulted in a hazard ratio (HR) of 0.64 [95%CI, 0.48–0.87] ( p = 0.019). Reasons for cup revision differed only by more cup revision due to loosening in CoC (26.2% vs.1 3.2%) ( p = 0.030). For aseptic loosening a revision rate of 0.153% [95% CI, 0.075–0.231] was seen in CoC and 0.058% [95%CI 0.019–0.097] in CoPE ( p = 0.007). Correction for head size resulted in a HR of 0.475 [95% CI, 0.197–1.141] ( p = 0.096). Incidence of different revision procedures did not differ between bearings. Over time the use of CoPE has increased and CoC decreased. Conclusions: A higher 2-year cup revision rate in press-fit THA was observed in CoC compared to CoPE. Cup loosening was the only significantly different reason for revision and seen more often in CoC and mostly aseptic. Future randomised controlled trials need to confirm causality, since the early cup revision data provided has the potential to be useful when choosing the bearing in press-fit THA, when combined with other factors like bone quality and patient and implant characteristics.