This paper explores the application of the deep Ritz method (DRM) to learn complex fracture processes within the phase-field modeling framework for brittle fracture. The authors address the challenges associated with approximating the energy landscape using neural networks (NNs) and the optimization of the energy functional. They discuss the design of NNs and training strategies to accurately capture crack initiation, propagation, kinking, branching, and coalescence. The DRM is compared with other physics-informed deep learning approaches, such as physics-informed neural networks (PINNs), and its advantages in handling non-convex energy functionals are highlighted. The method is applied to several benchmark problems, and the results are shown to be in qualitative and quantitative agreement with finite element (FEA) solutions. The robustness of the approach is tested by using NNs with different initializations. The paper concludes with a discussion on the computational cost and the potential for solving parametric phase-field fracture problems using the DRM.This paper explores the application of the deep Ritz method (DRM) to learn complex fracture processes within the phase-field modeling framework for brittle fracture. The authors address the challenges associated with approximating the energy landscape using neural networks (NNs) and the optimization of the energy functional. They discuss the design of NNs and training strategies to accurately capture crack initiation, propagation, kinking, branching, and coalescence. The DRM is compared with other physics-informed deep learning approaches, such as physics-informed neural networks (PINNs), and its advantages in handling non-convex energy functionals are highlighted. The method is applied to several benchmark problems, and the results are shown to be in qualitative and quantitative agreement with finite element (FEA) solutions. The robustness of the approach is tested by using NNs with different initializations. The paper concludes with a discussion on the computational cost and the potential for solving parametric phase-field fracture problems using the DRM.