Frequently Asked Questions

Differences between different AIGC models

Currently, we classify models into three categories based on their functionalities: text models, image models, and audio models, which correspond to generating textual, visual, and auditory content, respectively.

What are base models?

The differences between different models lie in their weights and network structures. If two models have the same network structure but different weights, they belong to the same base model. For lightweight fine-tuning models like LoRA, they are compatible only within the same base model.

Uploading model preview results

To facilitate viewing the model's performance, we require that a model's preview results be submitted when uploading a model.

Image generation models

If you use stable-diffusion-webui, by default, the generated images will include metadata about the parameters used to generate the image.

Text generation models

You need to upload a screenshot of the generated text, which we will convert to text using OCR. However, if there is other information in the image that interferes with the text, the result may not be accurate and may require manual input from you.

Audio generation models

You need to upload the generated audio file.