I have been playing with these for a couple of years. Out of maybe a dozen, one failed. The input impedance went low so the control voltage couldn't be set.
With the default control voltage of 2V the worst case error so far is about 4Hz.
From a cold start, the output is useless for about 3 minutes, and will settle within 15 minutes. With a calibrated control voltage it will be within 1Hz of 10MHz. There is significant drift for a day, and a little bit more for a week. After that they are reasonably stable.
I use them in experimental GPSDOs. I have not been able to get stability consistently below about 0.5mHz (i.e about 5 parts in 10^-11). I don't know if this is GPS, OCXO, environmental effects or bad design. Maybe a bit of all of them.
As part of a hobby GPSDO, they are excellent value.