Collocations in Arabic, especially in Qur'anic texts, play a significant role in constructing linguistic meaning and function. Although corpus linguistic studies of the Qur'ān have grown, research specifically analysing collocations of verbs and prepositions in this sacred text is limited. While previous studies have focused more on analysing the semantics of individual words or their morphological relationships, this study offers a novelty by integrating corpus-based collocation analysis in examining the relationship between the verb āmana ‘believe’ and its accompanying prepositions in Surah Al-Baqarah. This study uses a corpus linguistic approach to systematically identify collocation patterns with the help of AntConc software. The analysis includes frequency of occurrence, lexical distribution, as well as semantic interpretation of collocations in the context of the Qur'an. The results show that the collocation āmana bi ‘believe in’ is the most dominant form, appearing 18 times, which reflects the transcendental faith relationship between the subject and the object of belief, such as Allah or His revelation. Meanwhile, the collocations āmana li ‘believe for’ and āmana min ‘believe from’ reveal more complex social nuances and origins of faith, suggesting that the meaning of faith in the Qur'ān has wider contextual variations compared to conventional meanings. Thus, this study contributes to the study of corpus linguistics, Qur'anic exegesis, and Arabic language teaching, particularly in understanding how collocation patterns play a role in shaping theological meaning. The novelty of this research lies in the application of corpus-based collocation analysis in the study of sacred texts, which has rarely been done comprehensively in Arabic linguistic studies. The results of this study can also support a more accurate and contextualised translation of the Qur'an.