I think some of the discrepancies arise due to references to different IPCC assessment reports. The F-gas regulations use AR4 for calculations of CO2 equivalents. Under AR4, the GWP of R410A is 2088 and R32 is 675.
There is, however, a newer report, called AR5 (who would have guessed). This has reassessed the refrigerants' GWPs. So, just to confuse matters, some quote the figures from this report. R410A, for instance, is 1924 under AR5.
As Rob said, this document might help explain:
http://www.ior.org.uk/app/images/dow...%2030.5.14.pdf