How to Use DALL-E 3 in ChatGPT

DALL-E 3 is OpenAI's most advanced image generation model, integrated directly into ChatGPT Plus and Enterprise subscriptions. You can create detailed images from text descriptions without switching platforms or learning complex interfaces.

1. Verify your ChatGPT subscription. DALL-E 3 requires ChatGPT Plus ($20/month) or Enterprise access. Click your profile icon in the bottom-left corner and confirm your subscription status shows Plus or Enterprise. Free ChatGPT accounts cannot access DALL-E 3.
2. Start a new conversation with GPT-4. Click the model selector at the top of your chat interface and choose GPT-4. DALL-E 3 only works with GPT-4, not GPT-3.5. You'll see 'GPT-4' displayed prominently above the text input box when properly selected.
3. Write a detailed image prompt. Type your image description directly into the chat. Be specific about style, composition, colors, and mood. For example: 'Create a minimalist illustration of a coffee shop at sunrise with warm orange lighting and people reading newspapers.' DALL-E 3 works best with descriptive, detailed prompts.
4. Review the generated images. ChatGPT will display your images directly in the conversation thread within 30-60 seconds. DALL-E 3 generates one image per request, though ChatGPT may create multiple variations. Each image appears as a clickable thumbnail that expands to full size.
5. Download or modify your images. Click any generated image to view it full-size, then right-click and select 'Save image as' to download. To create variations, ask ChatGPT to 'make another version with different lighting' or specify exactly what you want changed.
6. Refine your results with follow-up requests. If the image isn't quite right, describe specific changes in your next message. Say 'make the background darker,' 'add more people,' or 'change this to a winter scene.' ChatGPT maintains context about your previous images and can build on them effectively.