With a slight trade-off in your final resolution, it is possible to reduce the effective pixel size of both datasets and process them together. This helped me merge two datasets collected on the same grid and microscope settings but with different pixel sizes. I don’t know if it works for data collected on two different detectors, but in my case, I could merge and improve the final map resolution by this method and sharing the same here:
Another way is to change one dataset (the smaller pixel size) using MotionCor3 which allows arbitrary -FtBin parameter (like 1.2717391304). Compared to the way @prafful explained the error is smaller because the spectral resolution of the micrographs is very fine, and Nyquist doesn’t need to be sacrificed as much, but it takes longer and involves different software. Doubt there is a difference in practice unless Nyquist is a serious concern.