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.
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.
Yes exactly, I am hoping that when the AMD Ryzen Halo gets wider release and more consumers have these 128GB devices, the market will justify a wider range of quantized model sizes. And then when I win the lottery and can buy one I'll have lots of nice options!
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.
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.
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?
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.
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.
Qwen Flash Next 3.8 … even at 3 bit quant it is very solid.
The market is too small.
Yes exactly, I am hoping that when the AMD Ryzen Halo gets wider release and more consumers have these 128GB devices, the market will justify a wider range of quantized model sizes. And then when I win the lottery and can buy one I'll have lots of nice options!
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.
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.
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?
Right i forgot. That's even worse, it's a form of data poisoning but it poisons humans and not ai.
It was, as insane as it sounds. They walked that back AFAIK.
Fable will still dump you back into Opus at the drop of a hat, but at least it will tell you when it happens.