While he won't disclose methodology at all, this is a very interesting claim given that I have found SUC to be ... typically ... believable.
---
RT @SmartUACat
Everyone is curious about Ukrainian losses. I don't know. Very limited circle of people know exact numbers. But I decided to find a way to estimate those.

So. It's around 15K.

While Russian losses reported by the GS is the lower estimate. I've confirmation on this one.
https://twitter.com/SmartUACat/status/1618595486331727880