This study evaluated the fracture resistance and marginal fit of CAD/CAM ceramic and composite inlays. Molars (N = 80) were prepared to receive Mesio-occlusal-distal (MOD) inlays and randomly divided into four groups to be restored depending on the materials: (a) HLD: heat-pressed lithium disilicate ceramic (IPS e.max Press), (b) CLD: CAD/CAM-fabricated lithium disilicate ceramic (IPS e.max CAD), (c) NC: CAD/CAM nano-ceramic resin (Lava Ultimate), (d) RC: Indirect resin composite (Filtek P60). Each group was randomly divided into two subgroups regarding the resin cement: (a) High-viscosity resin cement (Syntac, Variolink II), (b) Self-adhesive low-viscosity cement (RelyX Ultimate). After marginal gap and cement thickness measurements, specimens were loaded to fracture in a Universal Testing Machine (1 mm/min). Intact molars acted as the control group (n = 10). Data were analyzed using one-way and two-way ANOVA, Tukey’s tests (α = 0.05). Before cementation, CLD group showed significantly lower mean marginal gap (65 ± 22.4 μm) and after cementation, cement thickness was again the lowest with CLD (82.6 ± 24.6 μm) and the highest with HLD (108.4 ± 21.3 μm) (p < 0.001). The mean marginal gaps of inlays at the gingival margin were significantly higher than at the occlusal and the axial margins (p < 0.05). While material type significantly affected the mean fracture resistance (p < 0.001), the cement type had no effect on the results (p = 0.083). NC group (2486 ± 40 N) showed significantly higher mean fracture load compared to those of other three groups (1997.5 ± 60–2007 ± 30) (p < 0.05). The mean fracture resistance of control group with the intact teeth was significantly higher than those of all groups (p < 0.05) except for NC (p > 0.05).