Anatomy is one of the most important and difficult subjects in medical studies. Traditional teaching methods are limited by a lack of interaction and the inability to adapt to students’ diverse learning styles. That’s why we created a personalized chatbot AskAnato that uses the Retrieved Augmentation Generation technique to help students learn anatomy from a predetermined knowledge base. The particularity of our chatbot is his ability to enhance answers to queries with figures retrieved from the pedagogical anatomy website of the Faculty of Medicine and Pharmacy of Marrakech. To evaluate AskAnato’s effectiveness, various queries (n = 20) on the cardiovascular system were fed into AskAnato, ChatGPT 4.0-mini, and Microsoft Copilot. A panel of five experienced anatomists evaluated the responses for factual accuracy, relevance, completeness, coherence, and fluency using a 5-point Likert scale. A weighted score was introduced to fit the context of education. Overall performance of the three chatbots is comparable.