Abstract: Purpose: This work aims to create and validate a comprehensive set of test cases and 3D reference dosimetric datasets for model-based dose calculation algorithms (MBDCAs) in permanent implant prostate brachytherapy. These test cases address the limits of the standard TG-43 formalism towards improving dose evaluations in prostate cancer treatment, in accordance with the recommendations of the joint Task Group 186 of the American Association of Physicists in Medicine (AAPM), the European Society for Radiotherapy and Oncology (ESTRO), and the Australasian Brachytherapy Group (ABG). Acquisition and Validation Methods: Six test scenarios using the I-125 OncoSeed 6711 were simulated with three Monte Carlo codes: EGSnrc application egsbrachy with ebgui, MCNP with BrachyGuide, and PenRed. The test cases range from basic, e.g., single seed in water, to more complex configurations, e.g., 58 seeds in a virtual patient model for prostate treatment. The virtual model was developed from anonymized CT-based images, representing a real patient with and without calcifications. Dose comparisons were performed using both local and global metrics, with eb_gui as the reference for validation. For all test cases, the local agreement of Monte Carlo codes was within similar to 3.0% and the global agreement was 0.01%. Data Format and Usage Notes: The dataset, including Monte Carlo input files and other required information, is available online at https://doi.org/10.5281/zenodo.15282647. Potential Applications: The developed test cases provide an essential resource for the development, validation, and commissioning of MBDCAs. They have the potential to improve dosimetric accuracy in complex prostate implants while also providing a platform for future brachytherapy research.