OpenAI has released a major update to its image-making tool, now called ChatGPT Images 2.0. The upgrade aims to make the tool faster and much better at following specific instructions, which has been a common struggle for AI art programs in the past.
The new version is reportedly four times faster than the previous one. It promises to be much better at keeping things consistent. The new system is designed to lock in important details, allowing users to make specific edits like adding an object or changing a style while keeping the rest of the picture the same.
AI has famously struggled to spell words correctly inside images, often creating blurry or nonsensical symbols. OpenAI says this new model can now handle clear, readable text even in small sizes. This makes it a lot more useful for people trying to design things like posters or social media posts directly within the chat.
To make ChatGPT Images 2.0 easier to find, there is now a dedicated Images section in the ChatGPT sidebar. It includes simple presets and suggestions to help users who aren’t sure exactly what to type.
The update is now available starting for most users, usually to the paying subscribers.
