[puru]

Can transition metals show negative oxidation states like the one above?
Is OsO4 Covalent?

[movies]

Yes, transition metals can have a formal negative charge.  I'm not certain that your example is actually a real molecule, however.

OsO₄ is often drawn as a covalently bonded molecule, but I have been told that the metal-oxygen bonds are almost certainly ionic.  If you compare the electronegativities (2.2 for Os, 3.44 for O) then it's definitely ionic.

[AWK]

But, how you explain a boiling point of OsO4, which is 118 C.

[movies]

But, how you explain a boiling point of OsO4, which is 118 C.

I hadn't thought of that, but I would think that in the crystal phase the packing is not that of a typical ionic compound since there is a fairly dense layer of negative charge surrounding the positively charged core of the OsO₄.  This would result in a fairly large electrostatic repulsion between OsO₄ molecules.