Created by team Kernel ID on January 20, 2023 provides a solution to accomodate the need to generate good prompts to utilize the state of the art text to images technologies such as Stable Diffusion and DALL-E 3. generates prompts from long paragraphs for use in text to image models. have a potential application in generating concept arts from novels for adaptation to video games and animation. We used the j1-grande-instruct to summarize a given paragraph to a prompt which could be used to generate images with other machine learning models. The results showed that are able to generate a prompt discribing the scenery but still lacks the ability to describe the technical aspects of the image.

Category tags: