From 2011 to 2021, LAMOST has released a total of 76,167 quasar data. We try to search for gravitationally lensed QSOs by limiting coordinate differences and redshift differences of these QSOs. The name, brightness, spectrum, photometry and other information of each QSO will be visually checked carefully. Special attention should be paid to check whether there are groups of galaxies, gravitationally lensed arcs, Einstein crosses, or Einstein rings near the QSOs. Through careful selection, we select LAMOST J160603.01+290050.8 (A) and LAMOST J160602.81+290048.7 (B) as a candidate and perform an initial analysis. Components A and B are 336 apart and they display blue during photometric observations. The redshift values of components A and B are 0.2% different, their Gaia_g values are 1.3% different, and their ugriz values are 1.0% or less different. For the spectra covering from 3690 to 9100 Å, the emission lines of C ii, Mg, H γ, O iii, and H β are present for both components A and B and the ratio of flux(B) to flux(A) from LAMOST is basically a constant, around 2.2. However, no galaxies have been found between components A and B. Inada et al. identified them as binary quasars. But we accidentally find a galaxy group near components A and B. If the center of dark matter in the galaxy group is at the center between components A and B, components A and B are probably gravitationally lensed QSOs. We estimate that the Einstein mass is 1.46 × 1011M⊙ and the total mass of the lens is 1.34 × 1013M⊙. The deflection angle is 197 at positions A and B and the velocity dispersion is 261 km s−1. Theoretically, this candidate could be a pair of fold images of a strong lensing system by a galaxy group, and we will investigate the possibility when the redshifts of nearby galaxies are available.