Back to Live Demos, Openai shows that we can now chat with chat more visually. This means that you can ask for requests for images in a row and it will remember the context.

In this example, a photo of a coin was sent, and then the team asked Chatgpt to make it transparent, including requests.

Given that the improved image generation is already available or at least rolling out-Pakinomist’s editor-in-lawge, Lance Ulanoff, already tested the feature.

Lance took a selfie and uploaded it to chatgpt via the iPhone app. Then he asked for it to be transformed into anime style. The first time it gave him a full hair, but then corrected when he asked for it to be bald.

Sam Altman also teased that the original image generation model within 4O is designed to be a bit offensive within land if that’s what you are directing it for. The most important phrase that is “within reason” and no doubt many users will put it to the test.

Now the other demo asks for a colorful image describing the theory of relativity, with some extra humor. Altman also noted that the image generation model is a little slower, but that the result is much higher in quality.

In the other demo, the Openai team took a selfie and then asked chatgpt to make it an ‘anime style.’ It took several seconds, but it actually generated what was requested. You can see it above.

Sam Altman was then quick to note that the improved image generation is starting to roll out now in Chatgpt and Sora for Pro users, and it will also be available to free users.

We also see the process of the native image generation model within the 4o model, making the generated selfie an “agi -memory.”

Openai is debuting native image generation in the 4o model

And we’re on the way to the races – Sam Altman calls this one of the most fun progress, and it’s built -in image generation in the 4O model. He quickly noticed ‘It’s a big step forward’ and something that Openai has been excited about rolling out for a very long time, for a whole series of people.

Altman notes that the best way to explain it is to show it off, so we’re already in a demo. In just a few seconds after the prompt, Openai showed a picture with what the team said has ‘Perfect Text.’ Apparently, a leap shows in terms of understanding the prompt and creating the image with clear text and a unique point of view.

Openais Live Stream has begun, and in the construction of KL. 14 ET / 11:00 PT / 18:00 GMT Start time we are treated with different pictures. Some of these overlaps, but it updates every few seconds and shows all the different styles.

The live stream description notes that we hear from Sam Altman, Gabriel Goh, Praulla Dhariwal, Lu Liu, Allan Jabri and Mengchao Zhong discussing 4o image generation.

Well, the livestream title sheds much more light to see that we can expect … way more than the intitally teased image. It has the title “4o image generation in chatgpt and sora”, so that means we are likely to get improvements to create images within chatgpt and sora.

The mention of the latter may also mean more general improvements for text-to-video generation.

Under 15 minutes to go now!