This study investigates the mechanical properties of high-strength self-compacting concrete (HSSCC) through rigorous laboratory testing. Six input parameters—cement, water-cement ratio, nano-silica percentage, fine aggregate, coarse aggregate and temperature—are analysed with compressive, flexural, and split tensile strengths. The study explores the impact of elevated temperatures (100–800 °C) on high-strength self-compacting concrete (HSSCC). Results demonstrate a consistent decline in mechanical strengths, with nano-silica notably enhancing compressive strength, particularly at a water-to-cement (w/c) ratio of 0.30. Elevated nano-silica counters the decrease in flexural and split tensile strengths with rising temperatures. Mass loss shows a pronounced escalation between 400 and 800 °C, underscoring the importance of nano-silica for optimizing durability in diverse temperature conditions. A dataset of 135 data points is used to develop machine learning prediction models employing six algorithms. The hybrid model of XGBoost with Whale Optimization Algorithm (WOA) Regression emerges as the most accurate predictor, demonstrating strong generalizability. The chosen model exhibits exceptional predictive performance, as evidenced by high R2 values of 0.981 for compressive strength, 0.9842 for flexural strength, and 0.9898 for split tensile strength. Moreover, the model demonstrates precision with low errors, including standard deviations of 6.2, 2.1, and 3.2, and minimal mean squared errors, mean absolute errors, and mean absolute percentage errors. The Taylor diagram visually assesses model performance, with the hybrid model outperforming other models. This research contributes insights into optimising concrete strength prediction and provides a foundation for further exploration into additional influencing factors and admixtures. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2024.