Purpose: This study evaluated the accuracy of four implant-level impression techniques with optical microscopy on two angulated conical internal-connection implants. Materials and Methods: A master cast with two internal-connection implant analogs angulated 10 degrees from each other and a master framework were fabricated. Four impression techniques were examined: octagonal transfer impression coping, nonoctagonal transfer impression coping, nonoctagonal pickup impression coping, and nonoctagonal pickup impression copings joined together with autopolymerizing acrylic resin. Experimental casts were fabricated from 40 polyether impressions divided into four groups. After the master framework was seated on each of the casts, one abutment screw was tightened and marginal gap measurements between analog and abutment were recorded on the other side with an optical microscope. Results: The casts produced from nonoctagonal pickup impression techniques were more accurate than those produced using transfer impression techniques (P .05). However, there was no statistically significant difference in the accuracy between the unsplinted and splinted methods in pickup impression techniques (P > .05) or between the use of octagonal coping and nonoctagonal coping in transfer impression techniques (P > .05). Conclusion: The casts produced from nonoctagonal pickup impression techniques were more accurate than those produced by transfer impression techniques, regardless of whether they were splinted, for angulated conical internal-connection implants. Keywords: implant impression, implant-level impression, impression accuracy, internal connection