Follow me on X/Twitter for exclusive content: https://x.com/Citron_Legacy 😁
Check out all the Art styles I’ve made: https://civarchive.com/collections/10766871
Please support the artist by checking out their page🔞: https://www.pixiv.net/en/users/7964
I really like the way Zankuro draws pointy ears. Also this style reminds me of Disgaea!
Description
Trained on CivitAI Trainer
Images: 881
Epoch selected: 3/5
engine: kohya
unetLR: 0.0005
clipSkip: 1
loraType: lora
keepTokens: 0
networkDim: 4
numRepeats: 1
resolution: 512
lrScheduler: cosine_with_restarts
minSnrGamma: 5
noiseOffset: 0.1
targetSteps: 1102
enableBucket: true
networkAlpha: 16
optimizerType: AdamW8Bit
textEncoderLR: 0
maxTrainEpochs: 5
shuffleCaption: false
trainBatchSize: 4
flipAugmentation: false
lrSchedulerNumCycles: 3
FAQ
Comments (11)
Dude, your Lora is insane !
I wasn't expecting Flux to already have such good ones, it's so cool :D
Thanks for being my first poster!! (Let me give you a double buzz tip)
Thanks I really like this style. Zankuro was one of my first Pony Loras so I figured it was worthy of being one of my first Flux Loras.
Honestly your Zankuro Lora is a good way to compare the different AI models :)
I still remember messing around with the first one (I believe it was SD 1.5), image generation sure went a long way in terms of quality since ^^
Flux sure is fun and it works perfectly with Civitai's built-in generator, I can't wait to see what else you'll cook !
No good posts here, only evil, EVIL I SAY!
Posted one with and one without trigger word for comparision.
LOL It can still be a good post even if it depicts evil.
Thank you for the comparison! Its nice to see how much of a difference just the trigger made. (I'm glad I trained the trigger on this lol)
@CitronLegacy Yep, it definitely helped :)
I also made a zankuro Lora I hadn't posted, I think it's a bit more closer aesthetic wise but yours has better consistency, I'll try a merge and see what happens !
Zankuro has such a nice style I'm glad to here there are other fans.
I recommend generating images with both Loras at once to figure out the right ratio.
For example,
"<lora:CitronsLora:0.7>, <lora:SenkiAIsLora:0.7>, perfect pink hair cat girl"
This way you can figure out the perfect mix before merging.
Please post your results images/loras/whatever I'm very curious to see what you make!
@CitronLegacy yeah ofc, I'll post some! Excited to see how much it improves
You should post it
@22lava44 dont have it anymore, it wasnt that much of an improvement though compared to the new awesome models we have now.
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.