ChatGPT Images has been updated with a new flagship image generation model, providing users with more precise edits while maintaining details. The new model is 4x faster than the previous one, making it ideal for both practical edits and expressive reimaginings.
The new features in ChatGPT Images include:
* Enhanced editing capabilities, including adding, subtracting, combining, blending, and transposing elements
* Improved text rendering, capable of handling denser and smaller text
* More consistent preservation of branded logos and key visuals across edits
* 20% cheaper image inputs and outputs
With the new model, users can create images that better match their vision, from small edits to full reimaginings. The update also includes a dedicated home for Images in ChatGPT, featuring preset filters and prompts to spark inspiration.
The improvements in ChatGPT Image 1.5 have been evaluated using rerun examples from the initial image generation launch, showing clear progress across various cases. However, there is still room for improvement in future iterations.
Source: https://openai.com/index/new-chatgpt-images-is-here