Can you please create low noise model gguf?

#4
by dipper9876 - opened

On my system, when I use the original wan gguf + lightx2v lora, an iteration takes 14s (1280x720x81).
If I use the combined gguf that has embedded lora, an iteration takes 10s.

So... could you please create the low noise Q8_0 version as well, that would have the newest lightx low noise lora baked in? Thx!

Sign up or log in to comment