OpenAI has been in a giving mood lately, offering ever more features previously reserved for the pro version of ChatGPT to those on the free plan, and now it is the turn of AI image generation in the form of access to DALL-E 3.
DALL-E 3 is OpenAI's AI image generator and similar to tools like Midjourney, Flux or Ideogram it can create photorealistic images, render text and allow for inpainting.
With the launch of GPT-4o earlier this year OpenAI made most of its paid-for features including image analysis, coding and use of custom GPTs free for the first time. It even made GPT-4o, the most powerful model on the platform available to free users (with a daily usage limit) but image generation wasn't on that list.
How do I create images with ChatGPT for free?
It is straightforward to use, though, as ChatGPT does most of the work. As OpenAI wrote on X: "Just ask ChatGPT to create an image for a slide deck, personalize a card for a friend, or show you what something looks like" and it will make it for you.
Be as descriptive as you can or ChatGPT will fill in any gaps based on its own training date. So if you want the card to feature a cat, specify that fact. Also, text rendering isn't perfect yet but if you put the words in double quotes it will be more accurate.
While free users can now create images, they'll be restricted to two per day, so if you want to create multiple revisions or even enough images for a comic book, you'll need to find another platform or pay $20 per month for Pro.
What else is free in ChatGPT?
We’re rolling out the ability for ChatGPT Free users to create up to two images per day with DALL·E 3. Just ask ChatGPT to create an image for a slide deck, personalize a card for a friend, or show you what something looks like. pic.twitter.com/3csFTscA5IAugust 8, 2024
Earlier this year OpenAI announced its goal of putting the most powerful artificial intelligence technology in the hands of as many people as possible.
To achieve that goal the company gave free users many of the features Pro subscribers took for granted including the ability to use a range of custom chatbots and features created by third parties.
The biggest change was finally getting access to the most powerful model — in this case GPT-4o — rather than using the two-year-old GPT-3.5. That model has since been beaten in benchmarks by models that can run on a smartphone.
Free users also have access to upload functionality. This allows them to use the AI vision and analysis features of GPT-4o where it can look at the details of an uploaded image and outline what it contains. My favorite use for this is to give it ingredients and have it offer up a recipe.
