Question plays a significant role in the teaching-learning process. However, preparing the question manually is time-consuming. One type of question, Multiple Choice questions (MCQs) has recently become an active field of research. While studies in this field are not relatively new, they have predominantly focused on the English Language. Based on author’s literature study, there is a limited research dedicated to generating MCQs in the Indonesian language. In this study, we proposed an end-to-end Indonesian MCQ system along with exploring single-task & multi-task. It contains interconnected components: Question Generation (QG), and Distractor Generation (DG). The performance of these components was evaluated by BLEU, ROUGE-L, SBERT and an error analysis. Furthermore, the proposed MCQ results were further evaluated through 22 human respondents. The results show that the ROUGE-L score and relevancy percentage for Single-Task Question Generation and Multi-Task Question Generation are 37, 46% of 100 is relevant, and 19, 65% of 100 is relevant respectively, while the Single-Task Distractor Generator and Multi-Task Distractor All Generator achieve averages of 14.38, 32% of 300 is relevant and 14.36, 27.3 % of 300 is relevant. In addition, the Human evaluations of the best configuration MCQ yield scores of 1.38 / 5 for difficulty 3.99 / 5 for clarity and 2.07 / 5 for rigidity (Low the better).
