why this quant got weird repetition

[fusedtinker]

following this because I'm in the same boat.

That matches what I was seeing. I ended up in about the same place using SGLang. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing.

[vramscope]

Maybe, but I had the opposite result. The reported numbers can look compelling, but in ordinary usage I didn't see the same benefit.