EnvisionAI is a React-based application that provides users with a platform to explore the OpenAI API, specifically by experimenting with GPT-3 and GPT-3.5 text completion models. It also integrates AI freeware tools like leonardo.ai and Blockade Labs to reduce development time and costs, resulting in a more efficient, and visually stunning design process.
To install and run EnvisionAI locally, follow these steps:
- Clone the repository:
git clone https://github.com/Hafsa-Malik/EnvisionAI.gitor directly download the zip file. - Navigate to the project directory:
cd EnvisionAI - Rename the
.env.examplefile to.envand add your OpenAI API key. - Navigative to the server directory:
cd serverand install the dependencies:npm install - Start the server at port 3080:
node index.js - Open a new terminal and navigative to the client directory:
cd clientand install the dependencies:npm install - Go to
client/components/FeedbackSection.jsand add your emailjs serviceID, templateID and formID. - Start the development server:
npm start - Open your browser and visit
http://localhost:3000to access EnvisionAI.
You can customize the max_tokens and temperature values in the server/index.js file according to your preference. max_tokens determines the length of model's response message and temperature effects randomness of model, follow these steps:
- Open the
server/index.jsfile in your preferred text editor. - Locate the
app.post('/', async (req, res) => { ... })function. - Inside the function, you'll find the code snippet responsible for making the OpenAI API call. Modify the max_tokens and temperature values according to your preference. For example, to change max_tokens to 100 and temperature to 0.5. Increasing max_tokens will also increase the charges.
This project is licensed under the MIT License.