OpenAI patents image generation using row markers to speed up AI
OpenAI has filed a patent for a method that helps AI models create images more efficiently by adding special 'border tokens' at the start of each row of pixels. These markers allow the model to quickly locate specific parts of the image without checking every single pixel, which speeds up the process of generating visual content from text prompts.
In the filing’s own words
The present technology pertains to systems and methods for improving efficiency in multi-modal AI models, particularly when handling long token sequences. Aspects of the present technology facilitate attention across large token sets (e.g., making up an image) by using border tokens output by an AI model at the beginning of respective rows of a set of image tokens. Thus, the AI model can reference an image token's position based on its position from a border token. Further, aspects of the present technology facilitate training of multi-model AI models to reduce negative influence between modalities.
What exactly they claim (claim 1)
1. A method comprising: receiving, by an AI model of a generative response engine, a request comprising a prompt for generating an image; for a respective row of tokens in a set of rows of tokens making up the image, sampling, by the AI model, a border token to indicate a beginning of the respective row of tokens; attending, by the AI model, to at least one of the tokens making up the respective row by referencing the border token at the beginning of the respective row to locate the at least one of the tokens making up the respective row; generating, by the AI model, a subsequent token based, at least in part, on a spatial context obtained by referencing the at least one of the tokens making up the first respective row; and outputting, to a front end of the generative response engine, the image created from a set of image tokens generated by iteratively generating image tokens by the AI model using border tokens to reference image tokens within the set of rows of tokens.
Facts
- Company
- OpenAI (OpenAI OpCo, LLC.)
- Official title
- AUTOREGRESSIVE IMAGE GENERATION WITH BORDER TOKENS
- Publication number
- US20260289826A1
- Application number
- 19245903
- Filed
- Jun 23, 2025
- Published
- Sep 24, 2026
- Inventors
- 7
- Claims
- 20
Classified under
G06T 11/00 Physics & computingG06T 2211/441 Physics & computingMore from OpenAI in AI & software
- OpenAI patents voice system that detects when you stop talking Sep 22, 2026
- OpenAI patents system to catch AI models cheating on tasks Sep 10, 2026
- OpenAI patents method to check AI thinking for cheating Sep 10, 2026
- OpenAI patents system that turns photos into new AI-generated images Sep 10, 2026
- OpenAI patents AI that answers questions using only your company's secure files Sep 17, 2026
Summaries are simplified from the filing’s own abstract and first claim. A patent filing is not a product announcement.
Patent data and drawings: U.S. Patent and Trademark Office (USPTO). Nebula is not affiliated with or endorsed by the USPTO.