GPT Image 2.6: Flare vs Sunburst, Price and GIFs (Nov 2026)

GPT Image 2.6 is the image model OpenAI released on November 4, 2026. In ChatGPT, where it is called ChatGPT Images 2.6, it adds animated GIFs and saved reference images. In the API it comes as two models: gpt-image-2.6-flare, the smaller and faster one, and gpt-image-2.6-sunburst, the base model for detailed work and consistent edits. Both cost $8 per million image input tokens, $30 per million image output tokens and $5 per million text input tokens, the same token rates as GPT Image 2.5 and GPT Image 2. OpenAI says the new models generate images 50% faster than before.

Verification note: The facts on this page come from OpenAI's announcement, its API pricing page, the GPT Image 2.6 model page and the image generation guide, all read on November 6, 2026. The per-image dollar examples come from fal, a third-party API host. Kensa is an AI video generator: it does not run GPT Image 2.6, and we have not generated images with it ourselves. The last section covers turning a GPT Image 2.6 image into a video on Kensa, which we sell.

What's new in GPT Image 2.6

GPT Image 2.6 adds GIF output, saved reference images, markup-guided edits, SVG output and a complexity setting, split across ChatGPT and the API.

GPT Image 2.6 Flare vs Sunburst

Flare is the faster, smaller model, and Sunburst is the slower base model that holds fine detail and keeps edits consistent. They cost the same per token.

Pick Flare for drafts, batches of variations and simple layouts. Pick Sunburst when you are editing the same image over several rounds and a face, a product label or a logo has to stay exactly as it was.

Neither model works in the Responses API, Chat Completions or the Batch API, according to OpenAI's model page. GPT Image 2.5 and GPT Image 2 are in the Batch API at half price; GPT Image 2.6 is not, as of November 6, 2026.

How much does GPT Image 2.6 cost?

OpenAI bills GPT Image 2.6 by tokens, and fal's launch-day requests with Flare cost about $0.04–$0.05 per image at high quality and low complexity.

OpenAI API prices per million tokens, November 6, 2026:

OpenAI does not publish a per-image price table for GPT Image 2.6, and it notes that equal token rates don't mean an equal cost per image, because each model and quality setting uses a different number of tokens. The only per-image figures we found are fal's launch-day bills for Flare text-to-image at high quality and low complexity:

Sunburst and higher complexity settings use more tokens, so expect them to cost more per image. For scale, OpenAI's own table for the older GPT Image 2 puts a 1024 × 1024 image at $0.006 (low quality), $0.053 (medium) and $0.211 (high).

Is GPT Image 2.6 free?

In ChatGPT, OpenAI says Images 2.6 is rolling out on every plan, on web, desktop and mobile, and in Codex; it has not published usage limits for the Free plan. One news report says free accounts get GIFs in the coming weeks rather than at launch; we could not verify that. In the API there is no free tier: you pay the token rates above.

Sizes and limits

GPT Image 2.6 accepts any size whose sides are multiples of 16, up to 3,840 pixels on the long edge.

• Sides: both multiples of 16, neither longer than 3,840 px.
• Shape: no wider or taller than 3:1.
• Total pixels: 655,360 to 8,294,400 (4K). OpenAI calls anything above 2560 × 1440 experimental.
• GIF and SVG: no streaming or partial images. A GIF's frame count and frame size have to fit the same pixel budget, so cut frame_count or the size if a request is rejected.
• Video-ready sizes: 1536 × 864 is exactly 16:9 and 864 × 1536 is exactly 9:16. Both fit the rules above, so you can generate at the shape of your video instead of cropping later.

GPT Image 2.6 GIF vs an AI video clip

A GPT Image 2.6 GIF is a short silent loop of at most 16 frames. For a clip of several seconds with camera movement or sound, animate the image with an image-to-video model instead.

How to turn a GPT Image 2.6 image into a video

Generate the image at your video's shape, upload it to an image-to-video tool, describe one motion, and pick a model. On Kensa it takes four steps.

1. Generate at the right size and format. Ask GPT Image 2.6 for 1536 × 864 (16:9), 864 × 1536 (9:16) or 1024 × 1024 (1:1). Kensa accepts JPG, PNG or WEBP up to 10 MB. Large PNGs can go over that, so set output_format to jpeg or webp, or scale the image down.
2. Upload it to Image to Video. The image becomes the first frame. Grok Image Video and Seedance 2.5 follow your image's aspect ratio rather than the one you select. For the other models, pick the ratio that matches your image.
3. Describe one motion. One camera move and one subject action work best, for example "slow push-in, the cat turns its head toward the window".
4. Pick a model and generate. The credit price shows before you start, and if a generation fails, its credits are returned.

The smallest clip on each image-to-video model on Kensa, November 6, 2026:

Dollar figures use the Basic plan, $9.90 a month for 280 credits; see pricing for the other plans and one-time packs. New accounts that sign up with Google get 15 free credits, valid for 72 hours. That covers the smallest clip on Grok Image Video, Seedance 1.5 Pro, MiniMax H3 or Veo 3.1, but not on Seedance 2.5.

Illustrations, products, pets, landscapes and original characters animate best. Celebrities, children, brand logos and copyrighted characters aren't supported. Most models also reject realistic photos of people, and a photorealistic generated person can be rejected the same way. You're never charged for a rejection.

FAQ

What is GPT Image 2.6?

GPT Image 2.6 is OpenAI's image generation model released on November 4, 2026. It powers ChatGPT Images 2.6, which adds animated GIFs and saved reference images, and it is offered in the API as two models, gpt-image-2.6-flare and gpt-image-2.6-sunburst.

When was GPT Image 2.6 released?

OpenAI announced it on November 4, 2026. It rolled out in ChatGPT and Codex and reached the API the same day.

What is the difference between GPT Image 2.6 Flare and Sunburst?

Flare is the smaller, faster model: OpenAI says it can return an image in as little as six seconds. Sunburst is the base model for the most detailed work and keeps details more consistent across edits at medium or high quality. Both cost the same per token and support GIF, SVG, markup edits and the complexity setting.

How much does GPT Image 2.6 cost per image?

OpenAI charges $8 per million image input tokens, $30 per million image output tokens and $5 per million text input tokens, and it publishes no per-image table. On fal, launch-day Flare images at high quality and low complexity cost about $0.053 at 1024 × 1024 and $0.041 at 1536 × 1024. Sunburst and higher complexity use more tokens.

Is GPT Image 2.6 free?

In ChatGPT, OpenAI says it is rolling out on every plan, but it hasn't published Free-plan limits. The API has no free tier; you pay per token.

Can GPT Image 2.6 make videos?

No. It can make animated GIFs of 2 to 16 frames, with no sound. For a video clip, upload the image to an image-to-video model; on Kensa, clips run 4 to 30 seconds depending on the model, and the smallest Grok Image Video clip costs 5 credits.

Can I use GPT Image 2.6 in the Responses API?

No, not as of November 6, 2026. OpenAI's model page lists only the Image API: the image generation and image edit endpoints. It doesn't support Batch either.

Sources

• OpenAI: "Images, upgraded: introducing ChatGPT Images 2.6", November 4, 2026
• OpenAI API pricing, read November 6, 2026
• OpenAI model page: GPT-Image-2.6 Flare, read November 6, 2026
• OpenAI image generation guide (GIF, SVG, markup, complexity and size rules), read November 6, 2026
• fal: GPT Image 2.6, launch-day per-image costs
• TestingCatalog: OpenAI launches Images 2.6 with GIFs and image reference, November 5, 2026
• Kensa credits: kensa.cc/pricing, from Kensa's billing configuration, November 6, 2026