What models where?

[SoftologyPro]

Your readme links to 3 models

https://huggingface.co/google/gemma-2b
https://huggingface.co/black-forest-labs/FLUX.1-dev
https://huggingface.co/Alpha-VLLM/Lumina-Image-2.0

Which of these need to be downloaded?
Where should they be downloaded?
What directory should the --ckpt argument point to?

This is to use the gradio UI.

Thanks.

[ChinChyi]

@SoftologyPro
We apologize for the confusion. Actually, we use Gemma-2B as our text encoder and Flux VAE to encode images. You don’t need to worry about these two files, as the script will download them automatically. What you need to download is our weight file. Unfortunately, our diffuser is not yet available, but you can download the latest .pth format weight file from Google Drive. You can use it to run the Gradio demo or directly use our sample.py script for inference.

[SoftologyPro]

@SoftologyPro We apologize for the confusion. Actually, we use Gemma-2B as our text encoder and Flux VAE to encode images. You don’t need to worry about these two files, as the script will download them automatically. What you need to download is our weight file. Unfortunately, our diffuser is not yet available, but you can download the latest .pth format weight file from Google Drive. You can use it to run the Gradio demo or directly use our sample.py script for inference.

Download quota exceeded.
Maybe you can share it on Huggingface with a working script that uses those models?
Just a simple example of a command line Text-to-Image script would be ideal.
If you support Image-to-Image (ie feed a seed image in that is then modified by Lumina) I would also be interested in that.
Oh, and include a requirements.txt with package versions too.

[ChinChyi]

@SoftologyPro
Thank you for your suggestion. We have uploaded the .pth weight files to Huggingface, and you can directly specify the weight directory as the download directory. Additionally, the current version does not support image-to-image, but we plan to include it in a future release. Stay tuned!