Background: Historically, reduced-intensity conditioning (RIC) was recommended to be performed for older patients who were considered ineligible for myeloablative conditioning (MAC) before allogeneic hematopoietic stem cell transplantation (allo-HSCT). However, the evidence regarding the optimal conditioning intensity in younger patients with AML or MDS is weak and contradictory. Methods: PubMed, Medline, Embase, and other online sources were searched from the initial period to February 25, 2020. Odds ratios and 95% confidence intervals were calculated to estimate pooling effects. Results: Four randomized controlled trials (RCTs) about conditioning intensity involving 633 patients were included. There were no significant differences of 1/2/4/5 years progression-free survival (PFS) and relapse incidence (RI) between two conditioning intensities. Overall survival (OS) was similar at 1/2/4 years, but patients receiving RIC had a higher OS at 5 years. Additionally, RIC were associated with lower non-relapse mortality, less grade II-IV and grade III-IV acute graft-versus-host disease (GVHD), and lower incidence of chronic GVHD compared with MAC regimens. Subgroup analysis showed similar OS and RI for AML patients, and there was a trend towards lower NRM and grade II-IV aGVHD in RIC group. Available data for MDS indicated that OS, PFS, and RI were comparable. For intermediate-risk patients, there was no evidence that RIC is inferior to MAC. However, for high-risk patients, MAC tends to perform better. Conclusions: Based on the above results, it might be concluded that RIC is a feasible treatment option for adults with AML or MDS younger than 66 years, particularly those with intermediate-risk disease. Future RCTs incorporating of risk stratifications are warranted to guide the optimal decision under certain conditions.