Speed and Editing Update for OpenAI ChatGPT Images

Speed and Performance Improvements

OpenAI released a new update for the visual generation model ChatGPT Images. With the technical improvements made, the model is said to run up to four times faster compared to previous versions. This performance increase aims to boost efficiency especially in scenarios that require fast concept work and near-real-time visual generation.

Advanced Editing and Consistency

The update focuses on manipulation capabilities on existing visuals in addition to creating visuals from scratch. The new infrastructure minimizes the “loss of control” issue by preserving context in successive edits made on a visual. While light, composition, and core details are kept fixed, changes are made only in the desired areas.

The core technical capabilities offered by the model include:

• Adding, removing, and merging elements within a visual.
• Changing the position of existing objects.
• High accuracy in clothing, style, and atmosphere changes.
• Access via API with version GPT Image 1.5.

While the update is being rolled out globally, it was announced that comprehensive access for Business and Enterprise users will be provided in a later period.