Accurate and efficient fracture simulation is essential for structural design, integrity assessment, and life-cycle management. FRACSIM provides a unified environment that integrates finite element–based damage mechanics with artificial intelligence–driven modeling, streamlining practical engineering work-flows [1, 2]. The platform supports flexible geometry handling via parametric modeling and standard CAD import, coupled with rapid automated mesh generation and a comprehensive material library. It implements efficient algorithms for realistic reinforcement generation (including horizontal, vertical, random, and radial patterns) adapted to diverse structural configurations. High-performance interactive visualization, built on VTK.js, enables real-time exploration of geometry, meshes, materials, and scalar, vector, or tensor fields directly in the browser. FRACSIM is deployable as both a web and desktop application; the web edition (accessible at app.fracsim.nl ) leverages a Node.js stack and Firebase to provide secure, platform-independent, collaborative access. The FEM module employs a displacement-based gradient-enhanced damage model (GEDM) with spatio-temporal adaptivity. Independently, an AI module using physics-informed neural networks (PINNs) can be trained solely based on the governing equations or combined with FEM-generated or experimental data, and subsequently used for rapid, standalone predictions. The backend relies on DOLFINx[3] for FEM computations and NVIDIA PhysicsNeMo-Sym for AI, both extended with proprietary formulations to support advanced damage modeling workflows. Case studies demonstrate that FRACSIM successfully combines FEM accuracy with data-driven efficiency, establishing it as a flexible and extensible platform for physics-based and AI-based fracture simulation. Keywords: REFERENCES [1] Amani J., A transient-anisotropic gradient-enhanced damage model with displacement smoothing for failure analysis in quasi-brittle materials, PhD Dissertation, Delft University of Technology (TU Delft), 2023. [2] Amani J., Moeini R., Prediction of shear strength of reinforced concrete beams using adaptive neuro-fuzzy inference system and artificial neural network, Scientia Iranica, Vol 19 (2), pp. 242-248, 2012. [3] Baratta I.A., Dean J.P., Dokken J.S., Habera M., HALE J., Richardson C.N., Rognes M.E., Scroggs M.W., Sime N., Wells G.N., DOLFINx: the next generation FEniCS problem solving environment, , Preprint, 2023.