GPT Image 2.6 API: Sunburst vs Flare, GIF and SVG Output, Complexity and Pricing (2026)

Short answer

OpenAI released GPT Image 2.6 on November 4, 2026 (API changelog). It comes as two API models: gpt-image-2.6-sunburst for detailed visual work and gpt-image-2.6-flare for fast iteration. Both add a complexity setting, markup-guided edits, SVG output and animated GIF output, and both cost the same as GPT Image 2.5: $5 per million text input tokens, $8 per million image input tokens and $30 per million image output tokens. OpenAI says the two API models generate images 50% faster than before, and Images 2.6 is also live in ChatGPT and Codex across all tiers (@OpenAI, @OpenAIDevs).

Moyi API supports GPT Image 2.6: call gpt-image-2.6-sunburst or gpt-image-2.6-flare with one API key through the OpenAI Images API format. GPT Image 2.5 stays available on the same key. No OpenAI account, organization verification or card is needed — see how to use the GPT Image 2.6 API without them.

New accounts start with a $0 balance — top up before your first request. Top up from $5. If a request returns 402 (400 on the OpenAI format) with "wallet balance is insufficient", your balance is empty — your base URL and key are fine. Top up and retry.

Get an API key for GPT Image 2.6 — sign up and top up from $5; no OpenAI account needed.

What's new in GPT Image 2.6

All four features below come from OpenAI's image generation guide and work through the Images API (/v1/images/generations and /v1/images/edits):

• Complexity control. A new complexity request field sets how much the model plans before rendering: low, medium or high (the default). Lower complexity suits focused edits such as changing a background or a sign; higher complexity helps with dense text, complex layouts and relationships between objects, but can add latency and unwanted changes. It's separate from quality, and there's no extra fee for it.
• Markup-guided edits. On /v1/images/edits you can upload a markup overlay that circles or marks the area to change, and describe the change in the prompt. OpenAI added this for supported GPT Image 2.5 models as well. Markup-guided edits use the same token rates as other edits.
• SVG output. Set output_format: "svg" on /v1/images/generations to get an editable vector file, for icons and illustrations. It requires n: 1 and doesn't support streaming, partial images or output_compression; the edits endpoint doesn't return SVG.
• Animated GIF output. Set output_format: "gif" on /v1/images/generations and describe the motion and loop in the prompt. frame_count takes 2–64 frames (default 9) and frame_duration_ms takes 20–2000 ms per frame (default 200). The edits endpoint doesn't return GIFs.

A request body with the new fields, as described in OpenAI's guide:

{
  "model": "gpt-image-2.6-flare",
  "prompt": "A paper boat bobbing on small waves, seamless loop",
  "output_format": "gif",
  "frame_count": 12,
  "frame_duration_ms": 120,
  "complexity": "medium"
}

Sunburst vs Flare

OpenAI's launch post adds two concrete differences: at low quality Flare can return an image in as little as six seconds, and at medium or high quality Sunburst keeps details more consistent from one edit to the next (@OpenAIDevs).

A practical way to choose, the same as with 2.5: iterate on drafts with Flare, then switch the model string to Sunburst for the final image or for edits where details must stay put.

Official pricing

Per million tokens, from OpenAI's pricing page. Sunburst and Flare cost the same, and so does GPT Image 2.5:

Equal token rates don't mean equal cost per image: token usage depends on the model, size and quality. For SVG output you also pay for the reference image the API renders first. Moyi API's own prices are on the pricing page.

GPT Image 2.6 vs GPT Image 2.5

If you already call GPT Image 2.5, the upgrade is a model-name change. The 4K size rules we documented for 2.5 — explicit WIDTHxHEIGHT, longest side at most 3840, at most 8,294,400 pixels — are in the GPT Image 2.5 guide; check your sizes against the same limits when you move to 2.6.

Calling GPT Image 2.6 through Moyi API

The call is in the OpenAI Images API format, so the OpenAI SDK works with only base_url changed:

from openai import OpenAI

client = OpenAI(base_url="https://api.moyi-api.com/v1", api_key="sk-...")
result = client.images.generate(
    model="gpt-image-2.6-flare",
    prompt="A product photo of a ceramic mug on a wooden table, soft morning light",
    size="2048x2048",
)

Edits go to /v1/images/edits. The request parameters are the same as for GPT Image 2.5, documented for gpt-image-2.5-flare and gpt-image-2.5-sunburst, plus the 2.6 fields above. You don't need an OpenAI account: you top up a Moyi API balance from $5, by card, and the same key also works for GPT, Claude, Gemini and DeepSeek.

FAQ

When was GPT Image 2.6 released?

November 4, 2026, according to OpenAI's API changelog, as gpt-image-2.6-sunburst and gpt-image-2.6-flare.

Is GPT Image 2.6 more expensive than 2.5?

No. OpenAI lists the same token rates for both: $5 per million text input tokens, $8 per million image input tokens and $30 per million image output tokens.

Can GPT Image 2.6 make GIFs and SVGs?

Yes, on the generations endpoint: output_format: "gif" for animations (2–64 frames) and output_format: "svg" for vector files. Image edits don't return GIF or SVG.

Can I use GPT Image 2.6 on Moyi API?

Yes. Use gpt-image-2.6-sunburst or gpt-image-2.6-flare with your Moyi API key; GPT Image 2.5 is available on the same key.

Read more

• GPT Image 2.6 Flare API docs · GPT Image 2.6 Sunburst API docs
• GPT Image 2.6 vs Nano Banana 2.1, Nano Banana Pro, Seedream 5.0 and Qwen Image 3.0
• Use the GPT Image 2.6 API without OpenAI organization verification or a card
• GPT Image 2.5 API: Flare vs Sunburst, 4K sizes and pricing
• GPT Image 2.5 Flare API docs
• GPT Image 2.5 Sunburst API docs
• Model catalog