V2 is quantized in a better way and is 0.5 GB larger than the previous version.
On Hugging Face: huggingface.co/lllyasviel/flux1-dev-bnb-nf4
Submitted 3 months ago by Even_Adder@lemmy.dbzer0.com to stable_diffusion@lemmy.dbzer0.com
https://civitai.com/models/645429
V2 is quantized in a better way and is 0.5 GB larger than the previous version.
On Hugging Face: huggingface.co/lllyasviel/flux1-dev-bnb-nf4
m_f@midwest.social 3 months ago
Is there a good tl;dr on what is new and interesting about flux? Does it use a different architecture, or is it just bigger?
Even_Adder@lemmy.dbzer0.com 3 months ago
It can do text easily and usually gets hands right. The downsides are, it doesn’t know any styles, and people are just now starting to fine-tune it.