Evaluating the quality of test items is crucial for accurately assessing student competence in academic competitions such as physics olympiads. This study aims to analyze the quality of physics olympiad test items used in the Temu Fisika 2024 competition for senior high school students across North Sulawesi. A total of 40 multiple-choice items from the preliminary round of the Temu Fisika 2024 physics olympiad were analyzed. The responses of 35 high school students who participated in the competition were examined using the Rasch model through the QUEST program. The analysis covered item fit, reliability (item and person), difficulty level, and distractor effectiveness as complementary CTT analysis. The results showed that 33 items (82.5%) fit the Rasch model, while seven items showed misfit. The Rasch reliability indices were good (Item Reliability = 0.82, Person Reliability = 0.84). The difficulty levels were appropriately distributed, with 50% of items categorized as moderate, 22.5% easy, 20% difficult, 2.5% very easy, and 5% very difficult. However, distractor analysis revealed a significant shortcoming: only 15% of items had all distractors functioning effectively. This study is the first Rasch analysis of the Temu Fisika olympiad, providing empirical evidence that while the test is generally reliable, specific improvements are needed. The findings highlight the importance of revising misfitting items and non-functional distractors to improve the discriminating power of the test, particularly among high-ability participants. This study also provides a practical reference for improving item development and validation processes in academic competitions.