Delv
DALL-E
Getting Started Guide

How to Use DALL-E

A practical guide to get you up and running with DALL-E. Written by Delv Editorial, Delv Team.

Getting started with DALL-E

In this guide, you'll learn how to create and edit images using DALL-E by OpenAI. After following these steps, you'll be able to generate stunning visuals from your text descriptions and make edits to existing images.

Step 1: Sign up and set up

  1. Go to OpenAI's DALL-E page.
  2. Click on the "Sign Up" button in the top-right corner if you don’t already have an account. If you do, click "Log In".
  3. Follow the prompts to create your account or log in. You can access DALL-E for free with limited usage via Bing or opt for ChatGPT Plus for $20/month for more features.
  4. Once logged in, navigate to the DALL-E interface through the main dashboard.

Step 2: Your first image generation

  1. In the DALL-E interface, locate the text box labelled "Describe an image".
  2. Type in a clear and detailed description of the image you want to create. For example, "a cat wearing a wizard hat sitting on a stack of books".
  3. Click the "Generate" button. DALL-E will process your request and display several image options based on your description.
  4. Browse through the generated images. If you see one you like, click on it to view in full size and download.

Step 3: Get better results

  1. To improve the accuracy of your images, be specific with your descriptions. Include details about colours, styles, and settings.
  2. Experiment with different prompts. If the first results aren't what you expected, try rephrasing or adding more context.
  3. Use the "Edit" feature for inpainting. Click on an image, then select the "Edit" button to modify specific sections. Highlight the area you want to change and provide a new description for that part.

Pro tip

When generating images, use adjectives and visual elements in your prompts. For example, instead of "a dog", say "a fluffy golden retriever playing in a sunny park". This helps DALL-E create more vivid images.

Common mistake to avoid

Avoid vague descriptions. Phrases like "a nice scene" will yield unsatisfactory results. Always be as detailed as possible to get the best outcomes from DALL-E.