-
-
Join Trapezoid to create breathtaking NFTs in literally seconds.
Inspiration
Anyone who has generated art and images using generative AI models such a DALL-E-3 and Leonardo.ai is aware that lots of hard work goes into prompting when generating such images and art. What if there was a way to own your prompts and even sell them? After all, in the real world, artists sell their art, so what if you could sell digital art that you have created? This idea is what inspired us to create Trapezoid, a generative art prompt marketplace in which digital generative AI artists can sell their prompts as NFTs.
What it does
Trapezoid allows users to browse the marketplace, which is populated by NFTs other users have made. Users can purchase NFTs that interest them, and in return they receive the prompt that was used to construct the art, which can then be viewed in their portfolio. However, users can also create their NFTs and list them for sale in the marketplace. Users guide how the art will be constructed as they will be the ones prompting the generative AI models to create the art. If the first generation does not match their expectations, that is okay, as users can regenerate or refine their original prompt even more to reach the desired art before finally minting the NFT.
How we built it
The entire application was developed in ReactJs with NextJs. For the frontend of the application, HTML, CSS, and JavaScript were used, with the CSS being written in Styled Components for easier reuse of style definitions and to pass props in a more efficient way. The frontend was organized in such a way that it followed a modularized approach by leveraging dynamic components and centralized themes. This not only enhanced the maintainability of the platform, but also will facilitate the ease of updates and changes as Trapezoid will evolve. The backend of the application utilizes Thirdweb, Solidity, and Ethers. OpenAI was used to generate the images (the DALL-E-3 model was utilized).
Use cases
Trapezoid has the potential to be utilized in various creative and innovative ways. Here are some use cases for Trapezoid:
Digital art ownership: Artists can create and tokenize their generative art prompts as NFTs on Trapezoid, allowing them to sell and establish ownership of their creative prompts.
Artistic exploration: Users can experiment with different prompts to explore the capabilities of generative AI models, pushing the boundaries of creativity and generating unique digital artworks.
Learning about generative AI: Trapezoid can serve as an educational platform for learning about generative AI art. Users can study different prompts and the corresponding generated art to understand how AI interprets and transforms prompts.
What we learned
The most important thing we learned was how to integrate interdisciplinary fields into one collaborative platform. Planning out and developing Trapezoid taught our team how to integrate generative AI, blockchain technology, and web development all into one comprehensive application. During the planning phase, it was crucial for us to implement good practices concerning user-centric design so that the app can be used by people who are not even familiar with blockchain technology, NFTs, or cryptocurrency.
What's next for Trapezoid
The journey doesn't end here. We are planning to encrypt and decrypt the prompts used to create NFTs so there is an extra layer of protection for those who own the prompts, ensuring that no one other than the user can maliciously access them. We are also planning to include more generative AI models, such as Leonardo.ai, as this would give our users a more diverse and customized experience when constructing NFTs. Chainlink functions will also power the image generation process on-chain to fully decentralize the application (this utilization of Chainlink functions was started, but was not fully polished in time).
Log in or sign up for Devpost to join the conversation.