Are you about to post an Warhammer Lore Video on Youtube and need a thumbnail to post. Look no more. This Lora has been trained for this purpose!

Use landscape aspect ratio for guaranteed good results

Information about this LoRa:

This LoRa has been trained on thumbnails Warhammer 40K LoRe videos on SD3.5L base model using standard settings; DPM++_2M sampler with SGM_UNIFORM scheduler.

Training data used can be found here : https://imgur.com/gallery/TQdwvYA

The prompts are created using JoyCaptions: https://colab.research.google.com/github/camenduru/joy-caption-jupyter/blob/main/joy_caption_jupyter.ipynb

For best effect, take a reference image online, pass it through JoyCaption and use the output prompt with this LoRa. A finetuned SD3.5 Large model will give better results.

Keyword is "Warhammer 40K reaper portrait" . The term was suggested by the CLIP model from this notebook I've created: https://huggingface.co/datasets/codeShare/fusion-t2i-generator-data/blob/main/Google%20Colab%20Jupyter%20Notebooks/fusion_t2i_CLIP_interrogator.ipynb

You can download the training data for this LoRa as a zip file here. Feel free to use it for your own projects : https://huggingface.co/datasets/codeShare/lora-training-data/tree/main