Can't get any same result from Civitai
I can't replicate any of the prompts on CivitAI. I sue the correct vae file, check every setting and everyting is the same. For example with AOM3A3 I should get this:
But I have this:
All the generation parameters are copy-pasted:
masterpiece, best quality, ultra-detailed, illustration, 1girl, solo, outdoors, camping, night, mountains, nature, stars, moon, bonfire, tent, twin ponytails, green eyes, cheerful, happy, backpack, sleeping bag, camping stove, water bottle, mountain boots, gloves, sweater, hat, flashlight, forest, rocks, river, wood, smoke, shadows, contrast, clear sky, constellations, Milky Way, peaceful, serene, quiet, tranquil, remote, secluded, adventurous, exploration, escape, independence, survival, resourcefulness, challenge, perseverance, stamina, endurance, observation, intuition, adaptability, creativity, imagination, artistry, inspiration, beauty, awe, wonder, gratitude, appreciation, relaxation, enjoyment, rejuvenation, mindfulness, awareness, connection, harmony, balance, texture, detail, realism, depth, perspective, composition, color, light, shadow, reflection, refraction, tone, contrast, foreground, middle ground, background, naturalistic, figurative, representational, impressionistic, expressionistic, abstract, innovative, experimental, unique
Negative prompt: (worst quality, low quality:1.4), (realistic, lip, nose, tooth, rouge, lipstick, eyeshadow:1.0), (dusty sunbeams:1.0),, (abs, muscular, rib:1.0), (depth of field, bokeh, blurry:1.4), (greyscale, monochrome:1.0), text, title, logo, signature
Steps: 12, Sampler: DPM++ SDE Karras, CFG scale: 7, Seed: 457168307, Size: 768x512, Model hash: eb4099ba9c, Model: AbyssOrangeMix3 (AOM3A3), Denoising strength: 0.5, Clip skip: 2, ENSD: 31337, Hires upscale: 1.5, Hires steps: 12, Hires upscaler: SwinIR 4x
I think first picture are not using orangemix.vae, its too colorful, probably kl-anime
I'm getting the same image. I switched the VAE to kl-f8-anime2 based on the "Some Tips / Discussion" section and I got esentially the same image, just more saturated. The same section on the Civit AI page apologized: Sample image: I am terribly sorry! It seems that I triggered a bug related to the "DPM++ SDE Karras" sampler and batch generation while creating the sample image, which resulted in the inability to reproduce the sample image through the seed. -- So maybe this is why it doesn't work...
I think they referred to this bug, so I'm thinking maybe picking the "right" batch size (trial & error, anyone?) with the right A1111 version would actually generate the right sample image.
Oh well, My guess is the author didnt put the complete prompt, or the settings of the automatic1111 somewhat changed then.
Hello, i am searching for this model hash "Model hash: eb4099ba9c, Model: AbyssOrangeMix3 (AOM3A3)" - but it seems it doesn't exist anymore? I have the same name one, but they all have different hashes. idk what am I missing.