LORA
HiDream
Trained Words
Studio Ghibli style.
About this model
Used 's Ghibli dataset. 120 images, batch size 4, 100 epochs, 3000 steps. Learning rate: 1e-5. 13 hours running on 2 x rtx 5000 ada gpus. I used diffusion-pipeline since it allows distributing loads across multiple gpu's.
*Update V2.0 I doubled the lr rate to 2e-5 to make a more potent LORA.
Image2Image guidance: If you want to do img2img, I recommend setting sampler = lcm and scheduler = linear_quadratic. The denoise will be between .3 - .4 and steps around 5-10. Will need to adjust on a case by case basis.
Related Models
Similar AI models you may like
LORA
Pony
Not Artists Styles for Pony Diffusion V6 XL
⭐ 0.0
⬇ 648,297
LORA
ZImageTurbo
Hands XL + SD 1.5 + F1D + Pony + Illustrious + zit
⭐ 0.0
⬇ 287,527
LORA
Anima
Velvet's Mythic Fantasy Styles | Flux + Pony + illustrious
⭐ 0.0
⬇ 206,685
LORA
ZImageTurbo
NSFW MASTER
⭐ 0.0
⬇ 197,895
LORA
ZImageBase
Realistic Skin Texture style (Detailed Skin) XL + SD1.5 + F1D + Pony + Illu + ZIT + ZIB
⭐ 0.0
⬇ 160,424
LORA
NoobAI
Stabilizer IL/NAI/CK
⭐ 0.0
⬇ 148,490
LORA
Anima
Aesthetic Quality Modifiers - Masterpiece
⭐ 0.0
⬇ 122,033
LORA
ZImageBase
Detailed Perfection style (Hands + Feet + Face + Body + All in one) XL + F1D + SD1.5 + Pony + Illu + zit + zib
⭐ 0.0
⬇ 121,518