I have try with my 1060
This result give the same speed of Vanitigen, but it not support multi GPU at now.
Default settings are not optimal. Play with -p -b -t settings
And it doesn't check uncompressed until you say so.
For gtx1070 I got 80MK instead of 40MK with vanitygen.
But I dig for collisions also, not only 32BTC bounty.
While searching for 32BTC bounty alone I get about 240MK on gtx1070 and that's impressive!