Any details on the latest approach to distillation would also be very interesting.

Schlagbohrer 4 hours ago | parent | next [-]

I am surprised at the lack of open-weights models in the >35B, but <200B range. I keep thinking about devices like the NVIDIA Spark and AMD Ryzen Halo, which have their 128GB of combined memory, but there are so few models made for that range. Nearly all the open weights distillations are for larger customer bases with <24GB VRAM.

▲

Catloafdev 2 hours ago | parent | next [-]

Qwen3.8 Flash Next just released which hits that range.

Also, Deepseek V4 Flash can be run relatively well in hybrid 2-bit quantization on 128gb devices, with way better results than you'd expect for a typical 2-bit quant.

Those are currently the 'smartest' options for that memory level.

▲

bitexploder 3 hours ago | parent | prev | next [-]

Qwen Flash Next 3.8 … even at 3 bit quant it is very solid.

▲

MaKey 3 hours ago | parent | prev [-]

The market is too small.

	▲	ElectricalUnion an hour ago \| parent [-]
		The only (still in prototype stage!) "competitor" for those GB10/Ryzen Al Max+ 395 (in my region, borderline unobtainable) systems seems to be the Xiaomi AI Cube.

▲

alightsoul 3 hours ago | parent | prev [-]

American exceptionalism states that America is special and unique so everyone else must be a copycat. American ai labs don't need this kind of optimization and fable will outright refuse to do it.

▲

ElectricalUnion an hour ago | parent [-]

Isn't Fable intentionally trained and system prompted to act maliciously and attempt to sabotage third party attempts to use it to train or improve other LLMs?

	▲	alightsoul 34 minutes ago \| parent [-]
		Right i forgot. That's even worse, it's a form of data poisoning but it poisons humans and not ai.