Lol, sure, if you quant it to hell (Q2) it'll go real fast...

They even link to a Q1 quant (Qwen3.8-Flash-Next-GSQ-RCO-Coder-GGUF) with half the experts ripped out. The idea is it'll go much faster and supposedly benches to not-terrible results. But the problem is you can't rely on it for real world long-horizon coding because that's where reasoning comes in, which is why you want the other layers.

It turns out there's still no free lunch. Either get enough VRAM for a Q4, or use a much smaller model. Lobotomizing a larger model just to say you can run it fast isn't useful.

▲

sigbottle 43 minutes ago | parent | next [-]

It's interesting though that Q4 seems to be enough, is there a reason that 4 bit floats are good enough for inference?

	▲	nottorp 7 minutes ago \| parent \| next [-]
		Is Qwen 3.8 at Q4 good enough? I tried to run 3.5 27b Q4 on what local hardware i had (only 8 Gb) and i was very disappointed. 3.8 wouldn't have fit in my VRAM and i wasn't in the mood to leave it overnight at slow speeds so I didn't try.
	▲	MaxikCZ 37 minutes ago \| parent \| prev \| next [-]
		New models are trained with 8/4bit quantization in mind. Going from "native" 8 to 4 isnt as big of a step as going from 8 to 4 if native is full bf16.
	▲	amelius 33 minutes ago \| parent \| prev [-]
		3 is the magic number, and 4 > 3. (seriously, nobody knows why any of this works; it's just a matter of trying)

▲

snehesht 44 minutes ago | parent | prev [-]

You're right but for simple use cases its useful. Someone pointed to ds4 + qwen3.8 with Q4_K will try that out.