Dave got it, as usual. It's pretty much impossible to get an exact wavelength out of these diodes as they are made. With more complicated designs, you can get *some* wavelength tuning, and of course you can tune the bandgap some. There are so many variables that are so hard to control, you can get close and you can get an average around a certain number, but it's pretty much impossible to get it exact with how these diodes are made and used.
For instance, one of the main contributors to wavelength is the bandgap, which for violet lasers is controlled by indium incorporation into indium gallium nitride quantum wells. You literally have to control the number of indium atoms in the sea of gallium and nitrogen atoms. The number of atoms in the quantum wells will change the bandgap, and as you can imagine, it's REALLY hard to get an exact number of atoms into certain places. Each quantum well will only be a few nanometers thick, literally just a few dozen atoms thick. So a few more or a few less indium atoms will change the bandgap.
And then, even with the right bandgap, other things can shift the output. For instance, if the manufacturer uses a DBR mirror to increase the reflectivity of the mirrors, the construction of the DBR can change what wavelength lases and what doesn't within a few nm (it can shift the mode of a certain mode has a much better reflectivity than another mode).
