AI Catchup

OpenAI Launches ChatGPT Images 2.6 With GIFs and API Models

By 4 min read

OpenAI released ChatGPT Images 2.6 for ChatGPT and Codex, adding faster generation, more precise editing, and GIF creation. Developers can use GPT-Image-2.6 Flare and Sunburst through the API. OpenAI says generation is 50% faster; GIF creation is still rolling out, and the Help Center separately documents a reusable reference-photo workflow.

OpenAI released ChatGPT Images 2.6 for ChatGPT and Codex, adding faster generation, more precise editing, and GIF creation. Developers can use GPT-Image-2.6 Flare and GPT-Image-2.6 Sunburst through the API. OpenAI says overall image generation is 50% faster than before, while GIF creation is still rolling out in ChatGPT. (OpenAI's announcement; OpenAI Help Center)

Key Takeaways

Images 2.6 is an image-generation update across ChatGPT, Codex, and the API. The release adds GIF creation and more precise edits, names two API models, and reports faster generation. Separately, the Help Center says GIF creation is rolling out and describes how to save a reference photo for later use in a future chat. (OpenAI's announcement; OpenAI Help Center)

  • ChatGPT and Codex get Images 2.6. OpenAI says it is available across subscription tiers on web, desktop, and mobile. (OpenAI's announcement)
  • GIFs can be created from prompts or still images. GIF creation is rolling out, and ChatGPT cannot edit an animated GIF that you upload. (OpenAI Help Center)
  • The Help Center separately documents a reference-photo workflow. Its reference section explains how to save a photo and use it later, but gives no separate rollout, tier, or platform details for that workflow. (OpenAI Help Center)
  • The API adds Flare and Sunburst. Flare prioritizes speed; Sunburst targets higher quality and consistency across edits. (OpenAI's announcement)
  • OpenAI reports a 50% speed improvement. At low quality, OpenAI says Flare can generate an image in as little as six seconds. (OpenAI's announcement)

What Changed in ChatGPT and Codex

Images 2.6 brings faster generation and more precise editing to ChatGPT and Codex, and adds the ability to create animated GIFs. OpenAI says the release is available globally across subscription tiers on web, desktop, and mobile. The announcement also says image markup, already available in ChatGPT, comes to the API so developers can point to a specific part of an image when requesting an edit. (OpenAI's announcement)

For context, see our coverage of the earlier ChatGPT Images 2.5 release. The 2.6 announcement focuses on animation, editing control, and API availability, offering two API model options for speed and quality. (OpenAI's announcement)

GPT-Image-2.6 API Models

OpenAI lists two API models for Images 2.6. Flare is the speed-oriented option and can generate an image in as little as six seconds at low quality; Sunburst is aimed at higher-quality generation and greater consistency across edits at medium or high quality. OpenAI reports that Images 2.6 overall generates images 50% faster than before. (OpenAI's announcement)

Model name in the announcementBest fit stated by OpenAIVerified detail
GPT-Image-2.6 FlareFaster generation and professional visuals (OpenAI)At low quality, generation can take as little as six seconds (OpenAI)
GPT-Image-2.6 SunburstHigher-quality generation and consistent edits (OpenAI)Designed for medium- or high-quality image generation (OpenAI)

OpenAI's announcement does not provide API pricing in the release page. Check the live API pricing and model documentation before estimating costs rather than carrying over a rate from an earlier image-model version. (OpenAI's announcement)

GIF Rollout and a Separate Reference-Photo Guide

ChatGPT's GIF feature is still rolling out. The Help Center says you can ask ChatGPT to create a looping GIF or animate a still image, and you can revise a GIF ChatGPT made by describing the change in the conversation. It also states a specific limitation: ChatGPT cannot edit an animated GIF that you upload. (OpenAI Help Center)

Separately, OpenAI's Help Center describes a reusable reference-photo workflow: upload a photo, ask ChatGPT to save it with a name, and later call it in another chat using @ followed by that name. The reference-photo section does not give separate rollout, tier, or platform details for that workflow, even though the page lists general Images availability across tiers and platforms. The Help Center does not state a release-specific rollout for this workflow. (OpenAI Help Center)

Availability

OpenAI says Images 2.6 is available in ChatGPT and Codex across all subscription tiers on web, desktop, and mobile, while GPT-Image-2.6 Flare and Sunburst are available in the API. The Help Center separately says GIF creation is rolling out, so that capability may not appear in every ChatGPT account yet. (OpenAI's announcement; OpenAI Help Center)

Sources

Frequently Asked Questions

What is ChatGPT Images 2.6?

ChatGPT Images 2.6 is OpenAI's latest image-generation release for ChatGPT and Codex, with faster generation, more precise editing, GIF creation, and two matching API models. OpenAI says Images 2.6 generates images 50% faster than before.

Which GPT-Image-2.6 models are available through the API?

OpenAI lists GPT-Image-2.6 Flare and GPT-Image-2.6 Sunburst in the API. Flare is designed for faster generation, while Sunburst targets higher-quality generation and consistency across edits. Flare can generate a low-quality image in as little as six seconds.

How much faster is Images 2.6?

OpenAI says Images 2.6 generates images 50% faster than before. It also says Flare can generate an image in as little as six seconds when quality is set to low.

Can ChatGPT create and edit GIFs?

GIF creation is rolling out in ChatGPT. You can ask ChatGPT to create or animate an image and can revise a GIF it created, but ChatGPT cannot edit an animated GIF that you upload.

Are saved reference images available to every user?

OpenAI's Help Center describes a saved-reference workflow but gives no separate rollout, tier, or platform details for that feature. The page lists general Images availability across tiers and platforms for the product overall.

Get the weekly AI Catchup

Tools, practices, and what matters, in your inbox every week.