OpenAI introduces native Image Generator in GPT-4o

OpenAI has integrated an advanced Image Generator with a natively multimodal model in GPT-4o that’s capable of creating accurate and photorealistic outputs.

With the help of Image Generator, ChatGPT users can generate images via text and voice commands.

According to OpenAI, Image Generator excels at accurately rendering text, precisely following prompts, and leveraging GPT-4o’s knowledge base and chat context. Users can also upload images to use as inspiration or modify existing photos.

“These capabilities make it easier to create exactly the image you envision, helping you communicate more effectively through visuals and advancing image generation into a practical tool with precision and power,” OpenAI says in a blog post.

Because Image Generation is native to GPT-4o, users can refine images through natural conversation. It follows detailed prompts with attention to detail. And while other image generators struggle with 5 to 8 objects, GPT-4o can handle up to 20 different objects.

In addition, GPT‑4o can analyze and learn from user-uploaded images, seamlessly integrating their details into photorealistic pictures or other convincing image styles.

“Our model isn’t perfect. We’re aware of multiple limitations at the moment, which we will work to address through model improvements after the initial launch,” OpenAI admits.

According to the ChatGPT developer, cropping can be an issue with longer images. The Image Generator can also make up information, especially in low-context prompts. Requests to edit specific portions of an image generation are not always effective and may also alter other parts of the image.

OpenAI aims to maximize users’ creative freedom. However, ‘the bad stuff’ such as child sexual abuse materials (CSAM) and sexual deepfakes is being blocked. “When images of real people are in context, we have heightened restrictions regarding what kind of imagery can be created, with particularly robust safeguards around nudity and graphic violence,” OpenAI assures.

GPT-4o Image Generator is available immediately for Plus, Pro, Team, and Free users. The latter are limited to generating only a few images per day. The Image Generator will be available to Enterprise and Edu users ‘soon’. It’s also available to use in Sora.