A web application harnessing the power of the Real-ESRGAN AI model to upscale images by 2x. Simply select an image file, or drag and drop an image over the drop zone, and wait for the model to upscale the image. Once complete, compare and contrast the difference in quality using the slider, then download the image.

PixelAngelo is built with Next.js on the front end and Flask on the back end. When an image is uploaded, it is redirected to a Backblaze B2 storage bucket. The back end instructs a customized Runpod container to obtain the image from the B2 bucket, then utilize the RealESRGAN upscaling model to upscale the image by 2x with cloud-based GPUs. Once upscaled, the new image is returned to the B2 storage bucket where the front end fetches both original and upscaled images for display. All images are automatically deleted after 48 hours. Logo design created with Midjourney and customized with Gimp.

A web application that utilizes the OpenAI API and gpt-3.5-turbo model to create two or three-tone gradients based on user input. The generated gradients are accompanied by hex codes, a unique name, and a description for why the colors were chosen. Users can navigate between generated gradients and easily copy hex codes.

Gradient Wizard is built with React on the front end and Django on the back end. The user's input is sent through the OpenAI API to the GPT-3.5 Turbo model in prompt format. The API response returns hex codes, a unique name, and description in JSON format that dynamically updates the front end. The history of gradients is managed within the React component's state, allowing users to easily switch between gradients. Logo design created with Midjourney and customized with Gimp.