BACKGROUND: Paranoia, characterised by inaccurate fears that others intend to cause harm, can significantly affect social functioning. Research has demonstrated that paranoia exists on a spectrum of severity, with milder forms prevalent in the general population. The Revised-Green et al. Paranoid Thoughts Scale (R-GPTS) is the most commonly used measure of paranoia comprising a scale to assess ideas of reference and a scale to assess ideas of persecution. The aim of the study was to validate a German version of the R-GPTS and assess its psychometric properties in non-clinical and clinical groups. METHODS: This longitudinal study was conducted in Germany, including a non-clinical group (n = 601) recruited online and a clinical group of inpatients diagnosed with persistent depressive disorder (n = 102). Participants completed an online survey assessing paranoia, other psychotic experiences, depression, and anxiety. Statistical analyses included confirmatory factor analysis to evaluate the factor structure and measurement invariance across sex, time, and patient status. McDonald’s omega was estimated for internal consistency, and Spearman correlations for test-retest reliability, and convergent and discriminant validity. RESULTS: Confirmatory factor analysis supported the two-factor structure, solid evidence in favour of metric invariance for the R-GPTS A subscale and more mixed measurement invariance evidence for the R-GPTS B subscale. The German R-GPTS exhibited good-to-excellent internal consistency (McDonald’s omega : 0.87 to 0.92), test-retest reliability analyses showed moderate-to-strong stability over a 10-week period, and we observed evidence for convergent and discriminant validity. DISCUSSION: These findings suggest that the German version of the R-GPTS is a reliable and valid tool for assessing paranoid thoughts across various populations. However, caution is warranted when interpreting score differences, as measurement non-invariance may impact the comparability of results for the Persecutory Ideations subscale. Limitations include potential selection bias in the non-clinical group and a focus solely on persistent depressive disorder in the clinical group. CONCLUSIONS: This study confirms the psychometric robustness of the German R-GPTS and contributes to the understanding of paranoia assessment in diverse populations, highlighting the need for further research to explore its applicability across different psychiatric conditions. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12888-025-07538-0.