OpenAI Images 2.6 Adds GIFs, Reference Images

// 1h agoMODEL RELEASE

OpenAI Images 2.6 Adds GIFs, Reference Images

ChatGPT Images 2.6 adds native GIF generation from text prompts or reference photos, with playback-speed controls and frame-level editing. OpenAI is also bringing GPT-Image-2.6 Flare and Sunburst to the API, targeting faster generation and higher-quality output.

// ANALYSIS

The meaningful upgrade is workflow: ChatGPT is moving from static image creation toward lightweight, editable motion content.

–GIFs offer a practical middle ground between still images and full video generation.
–Reference-image support should improve consistency for branded assets, characters, and product visuals.
–Frame-level editing makes iteration more useful than regenerating an entire animation.
–Flare and Sunburst give API developers a clearer speed-versus-quality choice.
–Developers should wait for public API documentation before building around GIF output.

// TAGS

chatgpt-imagesimage-genmultimodalvisionapi

DISCOVERED

1h ago

2026-11-04

PUBLISHED

1h ago

2026-11-04

RELEVANCE

9/ 10

AUTHOR

OpenAI

// KEEP READING

More AI developer news from the feed

EXPLORE FULL FEED

MODEL43m ago

Hy4 adds vision, computer use for developers

Tencent’s Hy4 model adds image input and computer use, enabling agents to operate graphical interfaces, terminals, and office software. It is also trained for Unity, Unreal, Godot, and long-horizon productivity workflows, with Nous Portal offering 80% off for one week.

INFRA54m ago

OneAdmin Runs Hundreds of Computer-Use Agents

OneAdmin demonstrates a one-person company coordinating hundreds of AI agents through browser-based computer use. The agents operate across customer-support and onboarding workflows in parallel.

UPDATE59m ago

Tesseract Turns Images 2.6 Into Video

Mirage positions Tesseract as the agent-driven finishing layer for ChatGPT Images 2.6, turning generated visuals, SVGs, and storyboards into editable motion graphics or finished videos with animation, effects, and sound. Its native project model keeps layers, timing, and audio revisable.