Summary Objective To compare articular cartilage signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and thickness measurements on a 1.5 T and a 3.0 T magnetic resonance (MR) scanner using three-dimensional spoiled gradient recalled echo (3D-SPGR) and two 3D steady-state free precession (SSFP) sequences. Methods Both knees of five volunteers were scanned at 1.5 T and at 3.0 T using a transmit-receive quadrature extremity coil. Each examination consisted of a sagittal 3D-SPGR sequence, a sagittal fat suppressed 3D-SSFP (FS-SSFP) sequence, and a sagittal Dixon 3D-SSFP sequence. For quantitative analysis, we compared cartilage SNR and CNR efficiencies, as well as average cartilage thickness measurements. Results For 3D-SPGR, cartilage SNR efficiencies at 3.0 T increased compared to those at 1.5 T by a factor of 1.83 (range: 1.40–2.09). In comparison to 3D-SPGR, the SNR efficiency of FS-SSFP increased by a factor of 2.13 (range: 1.81–2.39) and for Dixon SSFP by a factor of 2.39 (range: 1.95–2.99). For 3D-SPGR, CNR efficiencies between cartilage and its surrounding tissue increased compared to those at 1.5 T by a factor of 2.12 (range: 1.75–2.47), for FS-SSFP by a factor 2.11 (range: 1.58–2.80) and for Dixon SSFP by a factor 2.39 (range 2.09–2.83). Average cartilage thicknesses of load bearing regions were not different at both field strengths or between sequences ( P > 0.05). Mean average cartilage thickness measured in all knees was 2.28 mm. Conclusion Articular cartilage imaging of the knee on a 3.0 T MR scanner shows increased SNR and CNR efficiencies compared to a 1.5 T scanner, where SSFP-based techniques show the highest increase in SNR and CNR efficiency. There was no difference between average cartilage thickness measurements performed at the 1.5 T and 3.0 T scanners or between the three different sequences.