Coronary lesions of intermediate severity often cause ischemia, and fractional flow reserve (FFR)-guided revascularization for these coronary lesions is safe and effective. FFR derived from coronary computed tomography (FFRCT ) is a noninvasive method for diagnosis of lesion-specific ischemia, but its performance for intermediate stenoses has not been examined to date. We examined the performance of FFRCT versus FFR at the time of invasive angiography in 66 vessels of 60 patients who were identified as having an intermediate stenosis, defined by quantitative coronary angiographic percent diameter stenosis 40% to 69%. Ischemia for FFRCT and FFR was defined as ≤0.80. Diagnostic performance of FFRCT was determined compared to an invasive FFR standard. Mean age of the study group was 63.5 ± 8.1 years (81% men). Thirty-one patients (47%) demonstrated ischemia with an FFR ≤0.80, with 2 of 16 (12.5%), 21 of 37 (56.8%), and 8 of 13 (61.5%) lesions of 40% to 49%, 50% to 59%, and 60% to 69% stenosis causal of ischemia, respectively. At an FFR ≤0.80 cutoff for lesion-specific ischemia, accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of FFRCT were 86.4%, 90.3%, 82.9%, 82.4%, and 90.6%, respectively, with an area under the receiver operator characteristics curve of 0.95 (p <0.001) and good correlation to FFR (0.60, p <0.0001). No biases between FFRCT and FFR were noted by Bland–Altman analysis (0.03 ± 0.12, p = 0.054). In conclusion, FFRCT is a novel noninvasive method for diagnosis of lesion-specific ischemia of coronary lesions of intermediate stenosis severity.