ChatGPT Images 2.6: GIF creation and what's new

Short answer: OpenAI announced Images 2.6 on November 4, 2026 for ChatGPT, Codex and the API. The headline change for most people is native GIF creation from a prompt or an uploaded image, with controls to adjust playback speed and share the result. OpenAI also says still-image generation is about 50% faster, and developers get GPT-Image-2.6 Flare (fast) and Sunburst (higher quality and consistency) in the API, plus image markup, SVG exports and a new complexity parameter. It is available globally today in ChatGPT and Codex across all subscription tiers on desktop, mobile and web.

If you already use ChatGPT for images, or you are choosing between ChatGPT, Midjourney, Canva, Adobe Firefly and Runway, this is the biggest consumer image feature OpenAI has shipped in weeks. Below we cover what OpenAI announced, how GIFs work, what the API adds, how it compares with alternatives, and what to check on safety and pricing. We worked from OpenAI's own blog post, ChatGPT release notes and Images 2.6 system card, and we say so where a detail comes from press coverage.

What OpenAI announced

In a post on OpenAI's blog dated November 4, 2026, OpenAI introduced Images 2.6 in ChatGPT, Codex and the API. The company says people now create 4 billion images per week across those surfaces — more than half a billion a day on average — and that Images 2.6 generates images 50% faster than before. Those speed and volume figures are OpenAI's claims.

The same day's ChatGPT release notes frame it as a new image generation model that "creates images faster and is better at preserving details across edits," for example keeping a character's outfit when you change the background. Alongside GIFs, the notes add saved reference images: you can save a photo under a name (OpenAI's example is a dog called Pepper) and reuse it in later chats with an @mention without uploading again.

Codex is OpenAI's coding agent in the ChatGPT desktop app. It is not listed as its own tool on Unloc today, so we mention it in prose only. For how ChatGPT and Codex plan usage will change later, see our earlier note on shared ChatGPT and Codex usage limits.

Native GIF creation in ChatGPT

GIFs are the consumer-facing headline. OpenAI's blog says you can bring motion to still ideas — animating a brand mascot or character, turning a product shot into a looping clip, or building reactions and storyboards before a full video shoot. The release notes say GIF creation is rolling out in ChatGPT: ask for a looping animation, or upload a sketch and ask ChatGPT to make the character wave. You can adjust the playback speed and share the GIF. The Verge reports you can also edit individual frames, and that before this update ChatGPT could not natively generate GIFs in a single step.

That matters for marketers, social teams and anyone who wants a short loop without opening a separate motion tool. It is still a GIF workflow, not a substitute for timeline video editors or generative video models. Treat OpenAI's OpenArt and Town customer quotes on the launch post as vendor case studies, not independent benchmarks.

Faster stills, Flare and Sunburst in the API

For developers, OpenAI offers two GPT-Image-2.6 variants in the API:

• Flare. Positioned for speed. OpenAI says that when quality is set to low, Flare can generate images in as little as six seconds, and that it improves layout quality for posters, product imagery and other professional visuals.
• Sunburst. Positioned for higher-quality generation and greater consistency across edits when quality is set to medium or high.

Images 2.6 also extends editing controls that ChatGPT already had. OpenAI says image markup comes to the API so teams can pinpoint where to change something, SVG exports make designs ready for websites, apps and print, and a new complexity parameter lets developers dial how much planning the model does — lower settings for quick edits, higher for scenes with many elements or constraints. OpenAI recommends low complexity for most editing tasks, paired with the image quality your workflow needs.

API pricing for Flare and Sunburst is on OpenAI's live pricing page. Check platform.openai.com/docs/pricing for current rates rather than relying on a screenshot from launch day.

Availability

OpenAI says Images 2.6 is available globally today in ChatGPT and Codex across all subscription tiers on desktop, mobile and web. GPT-Image-2.6 Flare and Sunburst are available in the API. The release notes describe GIF creation as "rolling out," so some accounts may see it slightly later than others even though the blog says it is available today.

How it compares with Midjourney, Canva, Firefly and Runway

ChatGPT Images 2.6 sits in a crowded image stack. The right comparison depends on whether you want chat-native stills and short loops, a dedicated image model, template design, or full motion/video.

• Midjourney. Still the reference for many people who care first about look and community workflows. Midjourney is not a general chat assistant with GIFs built into the same thread. If your job is exploration and aesthetics, Midjourney remains a strong specialist; if you want prompt-to-GIF next to writing and research in one chat, ChatGPT now has a clearer edge. See Midjourney vs Adobe Firefly and Midjourney vs DALL·E 2 for older image-model trade-offs (DALL·E naming on Unloc is historical; ChatGPT Images is OpenAI's current consumer path).
• Canva Magic Studio. Canva is strongest when the deliverable is a designed layout — social posts, decks, brand kits — with AI as one step inside a template editor. ChatGPT GIFs are faster for a one-off loop from chat; Canva is usually better when you need brand templates, team editing and export presets in the same place.
• Adobe Firefly. Firefly is aimed at people already in Photoshop, Express and other Adobe apps, with commercial-safety positioning that Adobe emphasises for brands. ChatGPT wins on convenience inside an assistant; Firefly wins when the next step is a layered PSD or an Adobe pipeline.
• Runway. Runway is a generative video and motion tool, not a GIF toggle on a still model. OpenAI's GIFs are short looping animations; Runway is for longer clips, camera moves and video editing workflows. If motion is the product, compare Runway (see Runway vs Kling) rather than treating ChatGPT GIFs as a video replacement.

In short: use ChatGPT Images 2.6 when you want stills and short loops without leaving chat; keep Midjourney or Firefly for specialist image taste or Adobe workflows; keep Canva for templated brand design; keep Runway (or similar) when you need real video, not a GIF.

Safety and provenance

OpenAI published a ChatGPT Images 2.6 system card on November 4, 2026. It says the core safety stack builds on Images 2.5, with additional safeguards for GIF creation, because motion can introduce risks that are less obvious in a single still — including sexualized movement and harmful meaning across frames. OpenAI also describes continued C2PA provenance metadata and SynthID watermarking through ChatGPT, Codex and the API. Treat the evaluation tables as OpenAI's own offline tests on adversarial prompts, not as a guarantee about production traffic.

What it means if you're choosing tools

For ChatGPT users, Images 2.6 is a practical upgrade: faster stills (per OpenAI), better edit consistency (per OpenAI), and GIFs without a separate app. For teams picking a stack, it narrows the gap between "assistant that can draw" and "tool that can move," but it does not replace Midjourney's image culture, Canva's design system, Firefly's Adobe integration or Runway's video pipeline. If you already pay for ChatGPT, try GIFs and the new edit behaviour on a real brand asset before buying another image seat. If image quality is your whole job, run the same brief through ChatGPT and your current specialist and compare.

FAQ

What is ChatGPT Images 2.6?

Images 2.6 is OpenAI's November 4, 2026 image update for ChatGPT, Codex and the API. OpenAI says it generates images about 50% faster, preserves details better across edits, adds native GIF creation in ChatGPT, and ships GPT-Image-2.6 Flare and Sunburst for API developers.

Can ChatGPT make GIFs now?

Yes. OpenAI's release notes say GIF creation is rolling out in ChatGPT: you can ask for a looping animation or upload a sketch and ask ChatGPT to animate it, then adjust playback speed and share. The Verge reports you can also edit individual frames. Availability may roll out gradually even though OpenAI's blog says it is available today.

What are GPT-Image-2.6 Flare and Sunburst?

They are the two Images 2.6 models in OpenAI's API. OpenAI positions Flare for speed (as little as six seconds at low quality, per OpenAI) and Sunburst for higher quality and consistency at medium or high quality settings. Check OpenAI's live pricing page for current API rates.

Is Images 2.6 available on free ChatGPT plans?

OpenAI says Images 2.6 is available globally in ChatGPT and Codex across all subscription tiers on desktop, mobile and web. Plan-specific image limits can still apply, so check what your account shows in ChatGPT.

How do ChatGPT GIFs compare with Runway or Midjourney?

ChatGPT GIFs are short looping animations made inside chat. Midjourney is a specialist still-image tool without the same chat-native GIF workflow. Runway is built for generative video and longer motion work. Use ChatGPT for quick loops next to writing and research; use Midjourney when still aesthetics come first; use Runway when you need video, not a GIF.