Flat Colour Anime!
A vibrant anime style trained using a high quality dataset captioned in natural language. This is my labour of love and I'm constantly working on improving it.
Please consider following to be notified on updates to this style and many more interesting styles I create ❤️
Important notes below! Please read
Keywords:
Check the version details on the right for the correct trigger words for your version. Currently these are:
"Flat colour anime style image showing" for Flux Dev and Schnell, and Z-Image Turbo
"fca_style" for Pony
"fca style" for SDXL
If you don't use the keyword, you won't get the most out of the style.
Version Key:
✅ Current
⛔️ Deprecated
Select the relevant version for your model from the top.
Update notes for version 3.4:
Now available for both Flux Dev and Schnell
Longer training time with a lower learning rate
Trained at 1024px vs 512px
Improved dataset quality
Added regularisation images to the dataset
Description
FAQ
Comments (16)
Dang, My system cant handle flux, but I've been seeing it become the next ponyxl of loras I'm scared I cant keep up anymore
The requirements for Flux are coming down quickly, hopefully you'll be able to run it soon :)
Have you tried to see if you can run the Flux NF4 version? It's more lightweight than even the FP8 variants.
@steamrick I havent tried it yet but I've been seeing alot of requirements far above my RTX 4060 8GBVRAM 32gb
@Jobee You may want to try NF4 in Forge, I think it should run even with 8GB VRAM. Schnell-NF4 can create pretty good gens with only 4 steps.
Hope it helps!
The next official pony model is being made on AuraFlow. So it will be much more accessible than FLUX while still being very advanced.
Just get a 16GB 4060TI for 500 euros or less. When the 50 series cards come out, prices will drop on the 40 series. You might be able to pick up a cheap, used 4080 then too. There might even be a dark horse 5060TI with 24GB of vram, who knows.
@OzzyOsman laptop.
@Jobee Oof.
@Jobee I run the NF4 version in Webui Forge on my 8 Gb VRAM 3060 Ti with 32 Gb system RAM. It takes over a minute to generate an image, but it works.
is this NF4?
No, FP8 or FP16. I don't know if it's NF4 is compatible with those Lora types, sorry.
@CrasH new update went out tonight on forge, loras work now
@SomebodyOnceToldMeh Oh awesome! Now if someone can get Loras working between different GGUF quantisations then we're golden!
@CrasH from my tests last night, it seems like your loras did work on the NF4 checkpoint, but it takes an insane amount of memory to load it.
NF4 Lora: 8gigs -> 16gigs (RAM)
vs
Medium Lora: 8gigs -> 31 gigs (RAM; and stall there for about 20 seconds before unloading and letting the GPU render. I do know that a new patch went out today that's suppose to help speed it up, but uhh yeah..)
@SomebodyOnceToldMeh Woah that's crazy. Hopefully someone can figure that out, I'm sure someone will.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.