Osegi
AI·1 min read

OpenAI Launches Images 2.6 With 50% Faster Generation and GIFs

OpenAI has released Images 2.6 in ChatGPT, Codex and the API. The company says images generate 50% faster, and the update adds GIF creation.

An AI image generation interface showing an animated GIF mascot next to model options for speed and quality.
Image: BRIDGE

OpenAI announced on Wednesday, November 4 (US time), that its new image generation version, Images 2.6, is rolling out in ChatGPT, Codex and the API. According to the company, images are now generated 50% faster than before.

In ChatGPT and Codex, the update is available worldwide on all pricing plans, across desktop, mobile and web. Developers using the API get two models: GPT-Image-2.6 Flare and GPT-Image-2.6 Sunburst.

Two models for speed or quality

Flare is built for speed. With quality set to low, it can produce an image in as little as six seconds. Sunburst targets higher quality when set to medium or high, and is designed to stay consistent across repeated edits instead of degrading. In ChatGPT, users pick a model from the model selection screen, while API developers set the reasoning effort to tune the speed of generation and editing.

GIFs, SVG and a new parameter

Images 2.6 can now create animated GIFs. OpenAI points to examples such as animating a brand mascot or character, looping a product photo, or turning a simple sketch into a moving character.

The API also gains image markup, which lets users point to the area they want changed and was already available in ChatGPT. Export to SVG is now supported, and a new complexity parameter controls how much the model plans before answering a request. Lower settings suit quick edits.

Why it matters

Faster generation and a clear split between a quick model and a quality-focused one give teams more control over cost and turnaround. GIF output and SVG export extend the tool toward marketing and design workflows, where animated assets and scalable graphics are routine.

Apps worth a look

SUMO: the number puzzle

Tap the right tiles, add them up, hit the target. Easy to learn, hard to master. Free, for all ages.

Tags:openaiimages 2.6chatgptimage generationgifapi

More from this day

See all →

More in: AI

See all →