There is an increasing amount of approaches to use chatbots to support teaching [1–5]. However, studies show that teachers and learning partners should cover empathic skills to achieve optimal learning progress [6]. Therefore, we investigated the extent to which widely used state-of-the-art LLM-based chatbots now cover the empathic skills that teachers and learning partners should have. To evaluate the chatbots’ empathy across various dimensions, we used established psychological questionnaires. Our results demonstrate that the analyzed chatbots perform comparable or higher than healthy humans in 86.15% of the analyzed empathy aspects, while they even outperform both genders in 27.69%. ChatGPT with GPT-3.5 achieves comparable or even higher scores than humans in 100% of the tested empathy aspects, exceeding the other chatbots. Our analysis shows that the investigated chatbots fulfill the requirements to enable a high level of motivation, a deep understanding of the material and increased satisfaction with the learning process.