Create images from text with OpenAI’s image synthesis tool. It uses latent diffusion to learn associations between words and images, which allows people to use word prompts to create images. Additionally, you can perform outpainting – to upload and extend the borders of an image, and visual bridging – to merge images together.
In the progression of developing machine learning, OpenAI has released its next model dubbed ChatGPT as a research preview. You can try out the program at chat.openai.com. It operates in a dialogue format that makes it possible to ask questions of the program and have it return follow-up questions, admit mistakes, challenge incorrect premises, and reject inappropriate requests. This model builds upon InstructGPT.
Read more about the release from the OpenAI blog post.