Verified: 2 years ago
SafeTensor
Type | |
Stats | 4,063 |
Reviews | |
Published | Mar 27, 2023 |
Base Model | |
Training | Steps: 10 Epochs: 20 |
Trigger Words | vectorstyle |
Hash | AutoV2 C9F7BB5118 |
1mb LORA trained on three select vector illustrations with jagged trees\mountains and slightly 'melting' look, using luisa_pinguin guide.
3 ways to trigger, with slight differences in outputs:
Main trigger: vectorstyle.
No trigger, just add LORA to prompt.
Secondary, a bit weaker trigger,: limited vector.
Refer to examples to see possible results, feel free to experiment. Works good in combination with my previous style LORA.
If you liked my work, you can donate here.
Trained on Anything V4.5 with anime.vae.pt and CLIP Skip 2.
Samples are 512x768\768x512, using OrangeCocoa 50/50 Mix , with OrangeMix VAE, and using EasyNegative embedding, CLIP Skip 2.