The application of ubiquitous technologies in the improvement of education strategies is called ubiquitous learning. This strategy amplifies the pedagogical potential of e-learning through a ubiquitous and contextualized perspective. On the other hand, a ubiquitous technological mediation in learning can also increase the isolation of learners and reduce the integration among colleagues. Strategies to encourage the group learning can minimize these possible side effects. In this sense, this article proposes UbiGroup, an agent-based model for ubiquitous recommendation of educational contents for groups of learners. UbiGroup aims to help teachers to search, select and distribute educational materials for groups. The model considers the group profile and the context where learners are. The recommendation for dynamic groups of learners through a consensus profile is the main scientific contribution of this research. The model was evaluated through simulated scenarios. The results were encouraging and show potential for implementing UbiGroup in real learning environments.