ObjectiveTo compare the implant–abutment connection microgap between computer‐aided design and computer‐aided manufacturing (CAD/CAM) milled or laser‐sintered cobalt–chrome custom abutments with or without ceramic veneering and titanium stock abutments with or without crown cementation.Material and MethodsSix groups of six abutments each were prepared: (1) CAD/CAM cobalt–chrome custom abutments: milled, milled with ceramic veneering, laser‐sintered, and laser‐sintered with ceramic veneering (four groups: MIL, MIL‐C, SIN, and SIN‐C, respectively) and (2) titanium stock abutments with or without zirconia crown cementation (two groups: STK and STK‐Z, respectively). Abutments were screwed to the implants by applying 30 Ncm torque. All 36 samples were sectioned along their long axes. The implant–abutment connection microgap was measured using scanning electron microscopy on the right and left sides of the connection at the upper, middle, and lower levels. Data were analyzed using the Kruskal–Wallis test (p < .05).ResultsMean values (μm) of the microgap were 0.54 ± 0.44 (STK), 0.55 ± 0.48 (STK‐Z), 1.53 ± 1.30 (MIL), 2.30 ± 2.2 (MIL‐C), 1.53 ± 1.37 (SIN), and 1.87 ± 1.8 (SIN‐C). Although significant differences were observed between the STK and STK‐Z groups and the other groups (p < .05), none were observed between the milled and laser‐sintered groups before or after ceramic veneering. The largest microgap was observed at the upper level in all groups.ConclusionsTitanium stock abutments provided a closer fit than cobalt–chrome custom abutments. Neither crown cementation nor ceramic veneering resulted in significant changes in the implant–abutment connection microgap.