TypeNFT

Type about the NFT you want to create.

Comment

Inspiration

Basically, we wanted to make something fun for this hackathon because it's the holiday season. So, we came up with the idea of making NFTs with AI-generated images. It's a fun project that you can show your family and friends, demonstrating how awesome AI and Blockchain are in a single project.

It became more exciting for us because we are using AI and Blockchain, the two most revolutionary technologies in recent times, together.

What it does

TypeNFT allows users, to turn their text inputs into unique and colorful NFTs (Non-Fungible Tokens). TypeNFT is a project built using a generative AI model and Blockchain. In this application:

  1. You can generate an image by typing/describing it.
  2. The image will be automatically transferred into an NFT and stored on the Flow blockchain.

  3. You can see the NFT you created.

    How we built it

    We built TypeNFT using React for the frontend, an AI text-to-image model to create the image, and the Flow blockchain to create NFTs.

  4. We used React to create a user-friendly, fun, and colorful frontend.

  5. We used the FLOW testnet to create the NFT, store it, and complete/test our web3 features.

  6. We used a text-to-image generative AI model from Hugging Face to implement the image generation feature.

    Challenges we ran into

    Like any other project, we had a lot of challenges while making this project:

  7. Understanding how to implement the NFT smart contract.

  8. Finding a good AI model that we can easily integrate into our project was hard.

  9. Storing the NFT in the blockchain was a tricky problem because the AI model outputs a blob (the size of a blob is huge to store in a blockchain, it would be very inefficient), so we stored the blob object in a URL and then stored the URL in the blockchain.

    Accomplishments that we're proud of

    This is our first time using an AI model in our project, although it was relatively easy; still, it was satisfying when we saw an AI-generated image pop up in the project for the first time.

We are proud of our design; it is simple. We wanted to make it look fun and colorful, and we think we achieved it.

What we learned

We learned how to implement an AI model in a project using the Hugging Face Inference API. We'll try to set up the model on our local machine to use it in our next projects.

Best use of AI and BLockchain

We use an AI model and the Flow blockchin together, the simplest and best way possible ( atleast we think so ) to create a project that you can use for showing others the fun and power of this technologies.

What's next for TypeNFT

We couldn't figure out how to store the image on decentralized storage (IPFS) so we can access it anywhere and deploy our project using Netlify.

That's our next step; we are trying to figure out how to use IPFS, and then after some refinement in the project, we will deploy it on the internet.

Built With

Share this project:

Updates