Double training images, down to single trigger word.










“I am Skel-o'-lantern, your friendly pumpkin! With a bright smile and mischievous charm, I’m here to light up your Halloween with laughter and joy. Oh please, don’t confuse me with that Jack Skellington, that guy is just bones.”
🎃Happy Halloween!🎃
V2: Double training images, down to single trigger word.
Usage Tips :
Trigger word: "Skel-o'-lantern"
The LoRA weight I have been using is 0.8, but feel free to try and experiment!
Generate setting recommended:
Sampler: best with eula a Simple, or dpmpp_2m Karras.
Steps: 30>
cfg: 7
Example prompts:
score_9, score_8_up, score_7_up, Skel-o'-lantern, surrounded by many jack-o'-lantern, no humans, sitting in graveyard, crossed legs, bat flying in the background, night, moon, clouds, starry sky, glowing eyes, glowing mouth,
V1: This is my first lora, train with less than 20 images, so it probably won't be very flexible and accurate.
Usage Tips :
You should use trigger tag "Skel-o'-lantern" with these two tag "skeleton", "jack-o'-lantern" to create accurate character lora.
The LoRA weight I have been using is 0.8, but feel free to try and experiment!
Generate setting recommended:
Sampler: best with eula a Simple, or dpmpp_2m Karras.
Steps: 30>
cfg: 7
Example prompts:
score_9, score_8_up, score_7_up, Skel-o'-lantern, skeleton, jack-o'-lantern, surrounded by many jack-o'-lantern, no humans, sitting in graveyard, crossed_legs, bat flying in the background, night, moon, clouds, starry sky, glowing eyes, glowing mouth,
Hope you all enjoy and have a happy Halloween!
Double training images, down to single trigger word.
1. The rights to reposted models belong to original creators.
2. Original creators should contact SeaArt.AI staff through official channels to claim their models. We are committed to protecting every creator's rights. Click to Claim
