Skip to content
Note—002DevelopingModels

ChatGPT Images 2.6 can animate a storyboard frame, but it is not video

ChatGPT now keeps a saved character or prop to call back in later chats, and makes short GIF loops for storyboards and motion tests. Each loop is one picture cut into frames, and at low quality the new developer models bill about four times as much per image.

Drafted by The Generated · Filed by The Generated

OpenAI's ChatGPT Images 2.6, released Nov 4, adds animated GIF loops and lets you save a picture of a character or prop, then call it back by name in later chats. For creators already spending credits on results they can't control, saved references are the everyday fix: one picture of a character or prop stays on hand across chats without uploading it again. The GIF loops add a cheap motion test, and the catch is a bigger bill for rough drafts made through OpenAI's developer access.

GIF creation is still rolling out in ChatGPT, according to OpenAI's release notes. Each loop is one picture cut into frames, so asking for more frames gets you smaller ones: one tester's 32-frame loop came back at 217 by 217 pixels. OpenAI's guide allows up to 64 frames, but the only paid outside test we found was refused above 16, and testers report choppy motion and flicker on longer actions.

OpenAI also released two developer models, Flare and Sunburst, and says they generate images 50% faster. Flare held that in outside tests, while Sunburst came out slower in one. The listed prices did not change, but at low quality each Flare image was billed about four times as much as on version 2.5. CellCog, a design service, found that, and a separate developer on OpenAI's forum posted the same multiple.

Our take: saved references fix a daily annoyance on their own, and animated storyboards are a real gift for anyone planning shots. But OpenAI kept the price list and moved the meter, and the creators who draft cheaply are paying for the extra speed.

Source: OpenAI, Nov 4, 2026; CellCog, Nov 5, 2026; The Verge, Nov 4, 2026.

Read the full article: thegenerated.com/stories/chatgpt-images-2-6-gifs-20261105

Provenance — OpenAI: Introducing Images 2.6; OpenAI on X; The Verge: ChatGPT will now make you animated GIFs; OpenAI API image prompting guide (GIF params) · Status: verified: HIGH (OpenAI + The Verge)
ChatGPT Images 2.6 makes GIF loops. Think storyboard, not video
Models

ChatGPT Images 2.6 can animate a storyboard frame. Just don't call it video

ChatGPT can now turn a frame into a short GIF loop and reuse saved reference images by name, which makes it a cheap motion test before video. The loops are sprite sheets, the documented 64 frames failed above 16 in one paid test, and at low and medium quality the new API models bill three to four times the tokens of 2.5.

Drafted by The Generated · Filed by The Generated
OpenArt co-founder and chief executive Coco Mao, smiling, with long dark hair and a gray sweater, photographed against a plain light gray studio background

A chibi office worker sits at a desk, watching a stock chart fall. Over six frames his eyes widen, his face crumples, tears stream down his cheeks, and the sixth frame slides back toward the first so the loop never ends. OpenArt made it for OpenAI's launch of Images 2.6 on Wednesday 4 November, and the prompt is printed under the clip. It doesn't ask for a video. It asks for "a single animation sprite sheet": exactly six frames, arranged in a three-column, two-row grid.

That detail explains most of what the update can and can't do for people who make films. Images 2.6 brings animated GIFs to ChatGPT, saved reference images you can call back by name in later chats, and two new models for the API (OpenAI's pay-per-use developer access), GPT-Image-2.6 Flare and GPT-Image-2.6 Sunburst, which OpenAI says generate images 50% faster than before. OpenAI's own pitch includes a line aimed squarely at filmmakers: creators can "use animated storyboards to explore character movement before video production." That is a pre-visualization pitch, and on the evidence from launch night it is the honest one.

Keep reading — 8 min

That’s all of Models.

5 Nov · 2 notes, filed as the record moved.

Models — archive

Also filed

AI made microdrama dubbing cheap, and the dubbing company's own books show who pays
AI dubbing made Chinese microdramas cheap to export, and Transn's own filing shows who pays

Transn says it localized 20,000+ Chinese microdramas with cloned voices, but its filing shows margins falling as AI cuts prices per order. Most titles were likely subtitled, and voice consent terms are unpublished.

China's top AI drama is a grassroots hit made entirely with ByteDance tools
A part-time crew made China's top AI drama, and every layer of it belongs to ByteDance

Into the Song has run ten 100-episode seasons on Hongguo since August. The viral debt story is wrong, the play counts belong to different counters, and the money figures are the team's own word.

Hyundai India is paying for a Diwali microdrama out of its social budget
Hyundai India is paying for a Diwali microdrama out of its social budget

Hyundai Motor India is funding a six-episode Diwali microdrama with Terribly Tiny Tales from its social budget and will judge it over 15 days. The budget and view counts are undisclosed, and nothing suggests AI was used.