Intrusion Detection Systems (IDS) play a crucial role in safeguarding network security by identifying and mitigating malicious activities. Distributed Denial of Service (DDoS) attacks remain a persistent threat, causing significant disruptions to online services. This study enhances DDoS detection and classification by evaluating Transformer-CNN-FNN hybrid model for both binary and multi-class scenarios. Following extensive experimentation, the proposed model achieved 99.95% accuracy for binary classification using the CIC-IDS-Collection dataset and 95.62% for multi-class classification using the CICDDoS2019 dataset. The model integrates the Transformer’s ability to capture long-range dependencies, CNN’s strength in local pattern recognition, and FNN’s precision in classification. A comparative analysis with existing models shows that the Transformer-CNN-FNN outperforms other approaches across key metrics, including accuracy, precision, recall, and F1 score. These results demonstrate the potential of hybrid models to enhance IDS performance and strengthen cybersecurity defenses against complex DDoS attacks.