Purpose The objective of this study was to evaluate the marginal fit and the flexural resistance of nickel-chromium frameworks welded by different techniques, gas-torch, laser, and tungsten inert gas (TIG), compared with that of frameworks made via one-piece casting. Methods To evaluate the marginal fit, a master model was fabricated simulating four implants. Transfers and replicas were used to transfer the positions of the implants to the model, using a silicone matrix. The bars were waxed up and casted. Three assessments of misfit were performed for each implant using a stereomicroscope before and after welding in two predetermined regions, totaling six readings for each implant. To evaluate the flexural resistance, one group was made casting the specimens in one piece. Other 3 groups using gas-torch, laser, and TIG welding techniques were made after sectioned transversally. The data showed normal distribution and two-way ANOVA for marginal fit and one-way ANOVA for flexural resistance, and Tukey's posttest (α=0.05) was performed. Results For the marginal fit, the three welding methods presented similar results and were different from one-piece casting. For the flexural resistance, significant differences were observed among the studied groups (p < 0.001), and the one-piece group presented higher resistance compared to the three welding techniques. Conclusions Within the limits of this study, the three welding techniques yielded similar misfit results, and the laser and TIG techniques presented similar flexural resistance but lower than gas-torch and one-piece casting.