OpenAI on Tuesday unveiled a new image engine for ChatGPT that it says takes a key step forward in rendering text and handling more complex requests.
Why it matters: When a capable new model pops, it can lead to a viral moment and a surge in app downloads.
Driving the news: OpenAI said ChatGPT Images 2.0 supports a wide range of aspect ratios and comes in both standard and a "thinking" mode with built-in reasoning.
- All users will have access to the standard version of the image model, while the thinking mode is reserved for paid subscribers.
- The new models will also be available to developers via an API.
Yes, but: All that extra thinking can mean images take longer for ChatGPT to produce.
The big picture: The lead for any one model is often a short one, as new models catch up and often overtake the leader.
- Google scored significant attention last year when it launched Nano Banana.
- OpenAI had a big bump of its own a few months earlier when it debuted a more capable model, with Studio Ghibli-style images going particularly viral.
What they're saying: OpenAI said it can't predict what will take off this time, but expects the new model to produce its own round of viral images.
- "We believe that we are going to have another moment here," OpenAI product manager Adele Li said during a briefing with reporters.
More importantly, though, OpenAI says the new model is well suited to a wide range of professional work including ads, posters and mock-ups.
- "It's not just a tool for making beautiful pictures," Li said. "It's a creative assistant."
