Introduction Accurate activity quantification is important for radiation dose calculations in Nuclear Medicine. Quantification accuracy varies due to inherent limitations of imaging methods, object shape, size and background levels. The aim of this study was to evaluate the accuracy of activity quantification for Tc-99m and Au-198 with a SPECT/CT system using Monte Carlo (MC) simulations. Materials and methods Segmented CT images of each phantom setup were used as input for the MC simulations. Three phantom studies and a patient study were simulated with Tc-99m and Au-198 respectively. A radioactive concentration of 0.27 MBq/ml* was placed in all cylinders or spheres mentioned below. (i) Cylinder phantom study: Six different sized cylinders filled with radioactivity* were simulated in the water filled large cylinder phantom (LCP), with two levels of background activity. (ii) Sphere phantom study: Six different sized spheres filled with radioactivity* were positioned in the LCP and simulated as mentioned above. (iii) RSD thorax phantom study: Two spheres filled with radioactivity* were positioned in the RSD phantom with three different levels of organ and background activity. This was repeated for three different sphere sizes. (iv) Patient study: A patient study with two spherical lesions was simulated. Corrections for attenuation, scatter and collimator response were applied during image reconstruction. Using CT regions of interest, counts from reconstructed SPECT images were converted to activity using a partial volume correction (PVC) and a calibration factor. Results For cylindrical sources, good accuracies were observed ( ⩽ 10% for Tc-99m and ⩽ 40% for Au-198). Similar results were obtained for the spherical sources in the LCP, RSD phantom and patient study ( ⩽ 10% for Tc-99m and ⩽ 20% for Au-198). Conclusion The accuracy of activity quantification varies with object size and from the results it is evident that a PVC remains necessary for the smallest sphere sizes.