[Haskell-cafe] 16 bit floating point data in Haskell?

Hi guys, Do we have anything like half precision floats in Haskell? Maybe in some non standard libraries? Or I have to use FFI + OpenEXR library to achieve this? Cheers, Oleksandr.