According to wikipedia, electrolysis has an efficiency of about 70%, with a theoretical limit of 94%. Cracking water will always require more energy than it will produce, i.e. there is no free lunch.

The point of electrolysis is that it could be used as "clean energy storage" for other clean energy (e.g., solar). Hydrogen gas can be moved long distances, unlike other storage options that are more stationary (like a hydroelectric dam), and has higher energy density (120 MJ/kg for what I can find online) then even the best batteries available (2.54 MJ/kg for Lithium-ion nanowire?).

It will always require more.

There was some very promising work done in the mid-1990s with sonoluminescence[1], but an apparent accident at the lab destroyed half of Chicago's industrial district, killing the project manager, who was the motivating force behind the project. Attempts to duplicate the work destroyed a government research facility in Leesburg, Virginia, and further efforts were abandoned as unviable. There was a documentary released soon after detailing the tragic costs of this scientific discovery and advancement.[2]

[1] https://en.wikipedia.org/wiki/Mechanism_of_sonoluminescence

[2] https://www.imdb.com/title/tt0115857/

No, it generates more! You just discovered the perpetuum mobile.