Even_Adder@lemmy.dbzer0.com to Stable Diffusion@lemmy.dbzer0.comEnglish · 3 months agoTowards Pony Diffusion V7, going with the flow.civitai.comexternal-linkmessage-square5fedilinkarrow-up121
arrow-up121external-linkTowards Pony Diffusion V7, going with the flow.civitai.comEven_Adder@lemmy.dbzer0.com to Stable Diffusion@lemmy.dbzer0.comEnglish · 3 months agomessage-square5fedilink
minus-squareScrubbles@poptalk.scrubbles.techlinkfedilinkEnglisharrow-up3·3 months agoI’ve never fully understood Pony. It’s something completely different? Not SDXL? Not SD at all? Flux I’m very curious to try. I have a pile of Loras right now, and curious if I’ll have to retrain them
minus-squareEven_Adder@lemmy.dbzer0.comOPlinkfedilinkEnglisharrow-up5·3 months agoPony is like an extreme fine-tune of SDXL. It’s the way it’s trained makes it incompatible with regular SDXL LoRA and Controlnets so it gets a different classification.
minus-squareamorangi@lemmy.nzlinkfedilinkEnglisharrow-up1·3 months agoCare to explain the fine tune, and what about it makes it incompatible?
minus-squareEven_Adder@lemmy.dbzer0.comOPlinkfedilinkEnglisharrow-up1·3 months agoThat’s all I know. I don’t really use the models and I don’t know any specifics.
I’ve never fully understood Pony. It’s something completely different? Not SDXL? Not SD at all?
Flux I’m very curious to try.
I have a pile of Loras right now, and curious if I’ll have to retrain them
Pony is like an extreme fine-tune of SDXL. It’s the way it’s trained makes it incompatible with regular SDXL LoRA and Controlnets so it gets a different classification.
Care to explain the fine tune, and what about it makes it incompatible?
That’s all I know. I don’t really use the models and I don’t know any specifics.