Pure Fine-tuned based on the flux.dev model,
the FP8 quantized version is more suitable for certain Asian-focused realistic model,
offering a film-like grain effect. Developed by Muchen and ARang.
When using this model,
please comply with the terms of Black Forest Labs:
License:flux-1-dev-non-commercial-license
https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md
If the picture is not good, please add the guide words: MUCHEN+ARang
Special thanks to https://www.houdeyun.cn/ for providing GPU computing power support.
Contact us:
Discord：https://discord.gg/YtmN6yWUpb
Email：[email protected]
our models: https://huggingface.co/Locolabs
Description
Pure Fine-tuned based on the flux.dev model,
the FP8 quantized version is more suitable for certain Asian-focused realistic model,
offering a film-like grain effect. Developed by Muchen and ARang.
When using this model,
please comply with the terms of Black Forest Labs:
License:flux-1-dev-non-commercial-license
https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md
Special thanks to https://www.houdeyun.cn/ for providing GPU computing power support.
Contact us: [email protected]
our models: https://huggingface.co/Locolabs
FAQ
Comments (2)
How many images, what was your training config, how many tags did you introduce, what were the loss finetune metrics, what were the target goals?
Pls keep training it till that ugly chin of flux is gone, this looks promising.