Objective: Jet diffusion flames are widely used in industry, energy, and power, including industry kiln stoves, power station boilers, and gas turbines. Flame stabilization is an important problem in jet diffusion flames because of its involvement in the safety of combustion equipment, combustion efficiency, and pollutant emissions. Investigating the characteristics of flame stabilization and the related fire safety issues is important for the design of practical combustion equipment. Many investigations into the stabilization characteristics of laminar jet diffusion flames have been conducted. Our understanding of the characteristics of the flame shape and stabilization behavior of laminar flames is clear. However, for transitional and turbulent flames, especially for lifted flames, under normal gravity conditions, the coupling of buoyancy, jet flow, and Kelvin-Helmholtz (K-H) instability at the flame edge makes the problem more complex. Methods: Experiments were conducted on laminar-to-turbulent lifted jet diffusion flames under normal gravity and microgravity conditions. Variations in the flame lift-off height and length during the transitional process and the stabilization behavior of "non-buoyant flames" were observed. This study analyzes and discusses the stabilization characteristics of the lifted jet diffusion flames under microgravity conditions based on the experimental data of flame lift-off height and flame length. Compared with the results obtained under normal gravity conditions, the influences of buoyancy on the characteristics of flame stabilization were further analyzed. Results: Results showed that lifted flames under microgravity and normal gravity conditions yielded similar critical Reynolds numbers corresponding to the start and end of the transitional stage, respectively. During the transitional process, the flame lengths under microgravity conditions are approximately twice those under normal gravity conditions. The flame lift-off heights under microgravity conditions are always lower than those under normal gravity conditions; however, the differences between normal gravity and microgravity decrease as the jet flow velocity increases. The flame lift-off height is significantly influenced by the jet upstream of the flame base. In the transitional stage, the jet flow exhibits intermittent breakup, causing flame lift-off heights to oscillate. The transitional and turbulent flames all exhibit severe separation phenomena, resulting in the variation of the flame length with time. Because the flame lengths under microgravity conditions are longer than those under normal gravity conditions, the development of the K-H instability at the flame edge toward the downstream region is greater, and flame length varies over a wider range. Conclusions: Although flame lift-off heights under microgravity conditions were relatively low, the flame lift-off height fluctuations under the two gravity conditions had a similar control mechanism. Flame splitting is somewhat random. As Re=2 460 (transitional regime), the mean separation frequencies under the two gravity conditions have only a slight difference. In the turbulent stage, as Re also increases, the mean separation frequency under microgravity conditions increases. However, in the turbulent regime, the flame separation frequencies under microgravity conditions are relatively lower, indicating that buoyancy can promote flame splitting. Moreover, the relationship between the Strouhal and Froude numbers for the flame-splitting phenomenon indicates that the jet Froude number cannot correlate with the separation/oscillation frequency of the flames under microgravity conditions.