The SDFR competition aimed to investigate the use of synthetic data for training face recognition models. It was held alongside the 18th IEEE International Conference on Automatic Face and Gesture Recognition (FG 2024) and involved two tasks. Task 1 required using a fixed face recognition backbone and limited dataset size, while Task 2 allowed more flexibility in model design and training. Participants used synthetic datasets, either existing or newly generated, to train face recognition models. The submissions were evaluated on seven benchmarking datasets, including LFW, CFP-FP, CPLFW, AgeDB-30, CALFW, IJB-B, and IJB-C. The competition highlighted the potential of synthetic data to improve face recognition performance, although there was still a gap in accuracy compared to real-world datasets. The competition also emphasized the importance of generating synthetic data that is diverse and representative across different demographic groups to reduce bias. The results showed that some submissions achieved competitive performance with existing synthetic datasets, but there was still room for improvement. The competition also discussed the challenges of generating synthetic data with sufficient intra- and inter-class variations and the need for further research in this area. The competition provided a platform for researchers to explore new methods for generating synthetic data and training face recognition models in a privacy-friendly manner.The SDFR competition aimed to investigate the use of synthetic data for training face recognition models. It was held alongside the 18th IEEE International Conference on Automatic Face and Gesture Recognition (FG 2024) and involved two tasks. Task 1 required using a fixed face recognition backbone and limited dataset size, while Task 2 allowed more flexibility in model design and training. Participants used synthetic datasets, either existing or newly generated, to train face recognition models. The submissions were evaluated on seven benchmarking datasets, including LFW, CFP-FP, CPLFW, AgeDB-30, CALFW, IJB-B, and IJB-C. The competition highlighted the potential of synthetic data to improve face recognition performance, although there was still a gap in accuracy compared to real-world datasets. The competition also emphasized the importance of generating synthetic data that is diverse and representative across different demographic groups to reduce bias. The results showed that some submissions achieved competitive performance with existing synthetic datasets, but there was still room for improvement. The competition also discussed the challenges of generating synthetic data with sufficient intra- and inter-class variations and the need for further research in this area. The competition provided a platform for researchers to explore new methods for generating synthetic data and training face recognition models in a privacy-friendly manner.