

qwen3.8-ridge-gguf runs great in ollama on the 9070xt in 4bit kv cache quant at 35t/s. You can also run deepseek 4.1 flash in strata if you have time. both models are very capable.


qwen3.8-ridge-gguf runs great in ollama on the 9070xt in 4bit kv cache quant at 35t/s. You can also run deepseek 4.1 flash in strata if you have time. both models are very capable.


enshitification always makes sense. Hear me out.


Apologies mean “I made a mistake and I learned from it so it won’t repeat.”
yeah enough humans don’t know that as well unfortunately. But yeah obviously LLMs don’t understand anything. That’s not how they work


You are confusing LLMs with diffusion models. LLMs generate text, not images. They can be used as inputs to diffusion models and are thus usually intertwined but are not responsible for generating the images themselves. I am not completely refuting your point in general. Generative models are capable of generalising to an extend, so it is possible that such a system would be able to generate such images without having seen them. But how anatomically correct that would be is an entirely different question and the way these companies vastly sweep through the internet makes it very possible that these images were part of the training
the 27b model fits completely on the gpu so ram and cpu makes no difference.
For the deepseek model you need a fast ssd because no matter how big your ram is that model won’t fit there since it is almost 300gb.
I’ve got a ryzen9 5900x and 40gb ram. I don’t know how much of an impact the cpu actually has because i think most of it is evaluated on the gpu but could be that the router that selects the experts is executed on the cpu.
in any case the main impact will probably be vram size, pci-e throughput and ssd read speed (and maybe cpu speed).