Type | |
Stats | 53 1,676 |
Reviews | (17) |
Published | Jan 5, 2025 |
Base Model | |
Training | Steps: 1,345 Epochs: 10 |
Usage Tips | Clip Skip: 2 |
Trigger Words | symphtia-style |
Hash | AutoV2 2DDE5F98D1 |
This LoRA is an experiment I did using images I created, as well as some generated on Suno, to develop a specific style. Feel free to experiment with the LoRA. Now using more random images as of v5.0. The only other site this is support on is on tensor.art.
Version 1 was created using previous AI-generated images of mine, along with images made by Suno, an AI music generator, that generates AI art as a cover. The same version 1 was trained on my ExoCore Merge Model, and (version 1) may not work with certain XL and Pony LoRAs and models. If you want to use V1 or V2 head on over to tensor all versions will be archived there.
Version 2 was made with more images, both generated by this LoRA and on the same AI music generator. Version 2, 3 and 4 were also trained on regular SDXL with more images in them each time.
Version 1 Trigger Words Below
Version 2 Trigger Words Below (Some strengths of the other triggers may be weak)
Version 3 Trigger Words Below (Some triggers may be weak)
Version 4 and 5 Trigger Words Below:
In Version 5 I added "Realistic Style" as a trigger.
Note: Removed the comparisons with trigger and no trigger from the description, check the showcase images for it.
After Version 5 I realized this LoRA became more of a quality LoRA ever since V3. For V6 I added on 63 images that I hope this will give it a style but if it doesn't it'll be more of a quality LoRA still than a style LoRA even if it was trained as a style LoRA.
⚠️ IF YOU'RE GOING TO STEAL MY MODELS, MAKE IT SO THE DOWNLOAD IS ENABLED. ⚠️