On Nov. 4, 2026, OpenAI published a post titled “Introducing Images 2.6.” OpenAI says it is introducing Images 2.6 in ChatGPT, Codex, and the API, with GIF creation in ChatGPT and the API alongside faster still-image generation and two new API models, GPT-Image-2.6 Flare and GPT-Image-2.6 Sunburst. For related coverage, see Canva Unveils Brush, Its Own AI Model for Editing Designs and Microsoft's MAI Image 2.7 Ranks No. 2 in AI Image Editing.
Key facts
| Item | Detail |
|---|---|
| Announced | Nov. 4, 2026 |
| Products | ChatGPT, Codex, OpenAI API |
| Headline features | GIF creation; image generation OpenAI says is 50% faster than before |
| API models | GPT-Image-2.6 Flare; GPT-Image-2.6 Sunburst |
| Scale (OpenAI figure) | About 4 billion images per week across ChatGPT, Codex, and the API; on average more than half a billion images per day |
| ChatGPT availability | Globally across all subscription tiers on desktop, mobile, and web |
| API pricing | See OpenAI’s pricing page (not stated in the announcement) |
Faster generation and the Flare and Sunburst models
OpenAI says Images 2.6 generates images 50% faster than before. For developers, the release adds two API models tuned for different speed-and-quality tradeoffs.
GPT-Image-2.6 Flare targets speed. When quality is set to low, OpenAI says Flare can generate images in as little as six seconds. OpenAI also says Flare improves layout quality for posters, product imagery, and other professional visuals.
GPT-Image-2.6 Sunburst targets higher-quality output. OpenAI describes Sunburst as offering higher-quality image generation and even greater image consistency across edits when quality is set to medium or high.
GIF creation in ChatGPT
OpenAI says GIFs bring motion to images: users can animate a brand mascot or character, turn a product shot into a loop, or even start from a simple sketch and watch it become an animated character.
OpenAI’s Nov. 4, 2026 ChatGPT release notes say GIF creation is rolling out in ChatGPT, with options to adjust playback speed and share the result. The same notes mention saved reference images so users can reuse a photo in later chats, citing an example like @Pepper.
API controls: markup, SVG, complexity, and quality
OpenAI says image markup, already available in ChatGPT, comes to the API, alongside SVG exports. The API also gains a new complexity parameter OpenAI ties to planning effort, while image quality controls visual detail. For most editing, OpenAI recommends setting complexity to low for faster edits.
Availability
OpenAI states Images 2.6 is available globally today in ChatGPT and Codex across all subscription tiers on desktop, mobile, and web. GPT-Image-2.6 Sunburst and Flare are available in the API; OpenAI points builders to its pricing page rather than listing rates in the announcement.
OpenAI links a system card covering safety evaluations and safeguards for the release, including GIF-related risk work.
What it means
Analysis (ours). OpenAI is treating still images and short motion as one surface area across ChatGPT, Codex, and paid API traffic, at a scale it quantifies in billions of images per week. Splitting Flare and Sunburst gives integrators an explicit latency-versus-fidelity knob rather than one default model, while consumer GIF tooling and saved reference images push reuse and brand-consistent loops inside chat. How aggressively GIF creation rolls out in every client, and how SVG plus markup land in production API workflows, will determine whether this reads as a creator feature or a broader design pipeline upgrade.
What to watch
GIF rollout completion in ChatGPT, uptake of saved reference images, Flare’s six-second low-quality claims in real workloads, Sunburst consistency on multi-step edits, and any follow-on API documentation for markup and SVG exports.