

Qwen 3.8 flash next on a 4090, using strata instead of llama.cpp, gives me about 45-55 tokens/sec. Mainly used for coding tasks. The model size is perfect. If the qwen 4.0 models can be a bit less wordy in reasoning but the same size and quality they’ll be really strong.







Jeffries is just fucking terrible.
I do not want to “get even” I want some GOD DAMN CONSEQUENCES for all the laws that were broken which at this point are way too numerous to list, but start with ICE’s flagrant civil rights violations, MURDERS, illegal detention and treatment of immigrants AND US Citizens. Consequences for Trump sacrificing US military personnel for his bullshit Iran war so he could trade energy stocks and make money off of their deaths.
Democratic leadership is absolute shit. This wave only happened despite them and because Trump is worse, but holy shit do we need to clean house.