This is the first LORA that I have trained for Ideogram 4. Trained on 30 images that were generated on the Ideogram.AI website. All images trained with JSON prompts. Let me know how it works for you!
Description
FAQ
Comments (2)
How did you achieve this clarity with the use of your LoRA?
I tried cfg3, euler, 28 steps and 4MP. But I don't achieve this quality. Did you run a second-pass uscaling?
Thank you for sharing your LoRA!
If you don’t mind, could you share your training settings? I’m trying to train a character/person LoRA with AI Toolkit, but I haven’t been getting good results so far. I’m using an RTX 5090.