OpenAI has released Images 2.6, now available in ChatGPT, Codex and via the API. The update focuses on faster image generation, new animation options such as GIF/sprite‑sheet output, and expanded controls for editing and exporting images.
Performance and usage figures
OpenAI says people create roughly 4 billion images per week across ChatGPT, Codex and the API — more than 500 million images per day on average. Images 2.6 is said to generate images 50% faster than the previous version.
For developers the API provides two model variants: GPT‑Image‑2.6 Flare and GPT‑Image‑2.6 Sunburst. Flare can produce images in as little as six seconds when quality is set to low, while Sunburst targets higher detail and improved consistency across edits when quality is set to medium or high. Flare also improves layout quality for posters, product imagery and other professional visuals.
Motion: GIFs and sprite sheets
Images 2.6 adds support for motion by making it straightforward to create looping GIFs or single‑file sprite sheets. OpenAI highlights use cases such as animating a brand mascot for campaigns, producing reaction GIFs and memes for social media, or creating animated storyboards to test character movement before video production.
The company shared example prompts that specify precise 3×2, six‑frame sprite sheets in a chibi toy‑like style, with strict constraints (same camera angle, identical frame dimensions, no captions or logos, and smooth looping). OpenAI also notes that a simple sketch can serve as the starting point for an expressive animated character useful in videos, games and other projects.
Editing controls and developer features
Images 2.6 gives developers more control over image creation, editing and downstream handoff. Key features include:
- Image markup in the API (previously available in ChatGPT), allowing teams to mark exactly where to make edits.
- SVG export to make designs ready for web, apps and print.
- A new complexity parameter to adjust how much planning the model does to fulfill a request — lower settings for quick edits, higher settings for complex scenes.
- Separate image quality controls that govern visual detail and fidelity; complexity defines how much planning or effort goes into the scene.
OpenAI recommends low complexity for most editing tasks to speed up edits while selecting the image quality appropriate to the workflow. For scenes with many constraints or elements, medium or high complexity is advised.
Early users, availability and safety
OpenAI cites Town, a startup building an AI work assistant, as an early access user of Images 2.6 in the API to explore how animated characters might add personality to everyday work interactions.
Images 2.6 is available globally today in ChatGPT and Codex across all subscription tiers on desktop, mobile and web. The GPT‑Image‑2.6 Sunburst and GPT‑Image‑2.6 Flare models are available via the API; OpenAI’s documentation provides pricing details. The company also points readers to safety evaluations and safeguards described in the system card for Images 2.6.
Why it matters
By combining faster generation, animation support and more granular editing controls, Images 2.6 aims to serve marketers, product teams and creative professionals who need both speed and consistency. The new developer features — image markup, SVG export and complexity/quality settings — are intended to make integration into production workflows smoother while maintaining OpenAI’s stated safety review processes.
