The aim of this study was to examine the relationship between rectus femoris muscle architecture and isokinetic concentric peak knee extension torque at various velocities. Twenty physically active men (age =21.1±1.41 years, weight =69.9±5.7 kg, height =176.4±7.07 cm) had their isokinetic concentric knee extensor (Biodex 4 Pro, Biodex Medical Inc, Shirley, USA) strength assessed at 60°/sec, 180°/sec and 300°/sec. Pennation angle, muscle thickness, and fascicle length of the rectus femoris was measured using real-time B-mode ultrasound (Logiq P5, GE Healthcare, UK). Pennation angle had a significant positive correlation to peak torque at 60°/sec (r=0.731, p=0.001), 180°/sec (r=0.802, p=0.001) and 300°/sec (r=0.685, p=0.001). There was a significant positive correlation between muscle thickness and peak torque at 60°/sec (r=0.718, p=0.001), 180°/sec (r=0.749, p=0.001) and 300°/sec (r=0.722, p=0.001). However, there was no significant correlation between fascicle length and the isokinetic peak torque values (p>0.05). In addition, pennation angle and muscle thickness were found to be significant contributors for predicting isokinetic knee extension torque (R2=0.47-0.64; p<0.01). Pennation angle and muscle thickness best predicted peak knee extension torque at 180°/sec (explained variance =64% and 56%, respectively). This study suggests that pennation angle and muscle thickness can be used to predict isokinetic knee extension torque in physically active men.