Image Generation with ChatGPT
ChatGPT now generates images directly using the GPT-4o model. Users can customize style and context, making the interaction creative and intuitive.
OpenAI has integrated image creation directly into ChatGPT using the GPT-4o model. This new feature allows for generating images that accurately follow user requests. ChatGPT can now also write text within images, enhancing user interaction.
The feature is available for all users, including Plus, Pro, Team, and Free. Previously, ChatGPT used DALL-E 3 to generate images, but now it leverages the same multimodal model to create graphic content. This means the entire process is smoother and more integrated.
Recently, I tested this new function by asking ChatGPT to create an image of Mario Draghi. After receiving some questions to contextualize the image, I specified that I wanted Draghi as a boxer. In just a few seconds, ChatGPT generated a surprising image of Draghi as a boxer, complete with gloves and a determined expression.
I then asked to add the text "Fight" on Draghi's shirt. Again, the image was ready in moments.
This feature is not just for fun; it can become a useful creative tool for those working with visual content, presentations, blogs, or social media.
Interacting with ChatGPT is natural and similar to working with a creative assistant. Users can choose the style, context, and details of the image. Additionally, it is possible to modify the generated image later. Thus, generating images with ChatGPT is easy and fun. Just express an idea, and the image comes to life instantly.
