GPT Image 2.6 Sunburst
Save 20%openai/gpt-image-2.6-sunburst
Generate PNG images and GIF memes, with 2K/4K output and token pricing.
GPT Image 2.6 Sunburst image & GIF examples
Real generated results (9): GIF memes, static memes or high-resolution images. Expand any example to copy the original prompt and request parameters.
Knock Knock, It’s Monday: claymation ghost
GIF · 512×512Auto-looping · 9 frames · 200 ms per frame
Prompt & generation settings
Create a square looping reaction GIF meme in handmade claymation stop-motion style: visible fingerprints in the clay, soft studio light, slight frame-to-frame jitter. An original small white clay ghost with round black eyes floats outside a wooden front door, knocks on it twice with a tiny hand, pauses and returns to the starting pose. Fixed camera, cozy evening porch. A large caption across the top reads exactly KNOCK KNOCK, IT'S MONDAY and stays perfectly still and readable in every frame. Seamless loop. No other text, logos or watermarks.
{
"model": "openai/gpt-image-2.6-sunburst",
"prompt": "Create a square looping reaction GIF meme in handmade claymation stop-motion style: visible fingerprints in the clay, soft studio light, slight frame-to-frame jitter. An original small white clay ghost with round black eyes floats outside a wooden front door, knocks on it twice with a tiny hand, pauses and returns to the starting pose. Fixed camera, cozy evening porch. A large caption across the top reads exactly KNOCK KNOCK, IT'S MONDAY and stays perfectly still and readable in every frame. Seamless loop. No other text, logos or watermarks.",
"size": "512x512",
"quality": "low",
"output_format": "gif",
"frame_count": 9,
"frame_duration_ms": 200
}
Wait, What?: 90s anime office worker and floating math
GIF · 512×512Auto-looping · 9 frames · 200 ms per frame
Prompt & generation settings
Create a square looping reaction GIF meme in a 1990s anime cel-shaded style with soft film grain. An original office worker in her thirties with long auburn hair tied in a low ponytail and round glasses sits at a desk in front of a laptop in a bright open-plan office. Simple white chalk-style math symbols (plus, multiply, pi, square root, equals) float and slowly rotate around her head while she stares at the screen in confusion, blinks twice and tilts her head slightly, then returns to the starting pose. Fixed camera, chest-up framing, clean consistent face in every frame. A large caption across the top reads exactly WAIT, WHAT? and stays perfectly still and readable in every frame. Seamless loop. No other text, logos or watermarks.
{
"model": "openai/gpt-image-2.6-sunburst",
"prompt": "Create a square looping reaction GIF meme in a 1990s anime cel-shaded style with soft film grain. An original office worker in her thirties with long auburn hair tied in a low ponytail and round glasses sits at a desk in front of a laptop in a bright open-plan office. Simple white chalk-style math symbols (plus, multiply, pi, square root, equals) float and slowly rotate around her head while she stares at the screen in confusion, blinks twice and tilts her head slightly, then returns to the starting pose. Fixed camera, chest-up framing, clean consistent face in every frame. A large caption across the top reads exactly WAIT, WHAT? and stays perfectly still and readable in every frame. Seamless loop. No other text, logos or watermarks.",
"size": "512x512",
"quality": "low",
"output_format": "gif",
"frame_count": 9,
"frame_duration_ms": 200
}
Stonks: 90s low-poly alien and a rising arrow
GIF · 512×512Auto-looping · 9 frames · 200 ms per frame
Prompt & generation settings
Create a square looping reaction GIF meme in a deliberately cheesy late-1990s low-poly 3D render style with flat shading and a glossy blue gradient backdrop. An original cheerful green alien in a navy business suit stands beside a big green arrow chart that zigzags upward; the arrow grows taller while the alien points at it proudly, then the arrow resets and the loop restarts. Fixed camera. A big glossy 3D caption across the top reads exactly STONKS and stays perfectly still and readable in every frame. Seamless loop. No other text, logos or watermarks.
{
"model": "openai/gpt-image-2.6-sunburst",
"prompt": "Create a square looping reaction GIF meme in a deliberately cheesy late-1990s low-poly 3D render style with flat shading and a glossy blue gradient backdrop. An original cheerful green alien in a navy business suit stands beside a big green arrow chart that zigzags upward; the arrow grows taller while the alien points at it proudly, then the arrow resets and the loop restarts. Fixed camera. A big glossy 3D caption across the top reads exactly STONKS and stays perfectly still and readable in every frame. Seamless loop. No other text, logos or watermarks.",
"size": "512x512",
"quality": "low",
"output_format": "gif",
"frame_count": 9,
"frame_duration_ms": 200
}
Still Waiting: lo-fi watercolor bear on a bench
GIF · 512×512Auto-looping · 9 frames · 200 ms per frame
Prompt & generation settings
Create a square looping reaction GIF meme in a soft lo-fi watercolor illustration style with visible paper texture and a muted autumn palette. An original small brown bear sits alone on a park bench at dusk holding a phone, checks the screen, sighs and looks up while a few leaves drift down, then returns to the starting pose. Fixed camera, quiet park. A handwritten-style caption across the top reads exactly STILL WAITING and stays perfectly still and readable in every frame. Gentle seamless loop. No other text, logos or watermarks.
{
"model": "openai/gpt-image-2.6-sunburst",
"prompt": "Create a square looping reaction GIF meme in a soft lo-fi watercolor illustration style with visible paper texture and a muted autumn palette. An original small brown bear sits alone on a park bench at dusk holding a phone, checks the screen, sighs and looks up while a few leaves drift down, then returns to the starting pose. Fixed camera, quiet park. A handwritten-style caption across the top reads exactly STILL WAITING and stays perfectly still and readable in every frame. Gentle seamless loop. No other text, logos or watermarks.",
"size": "512x512",
"quality": "low",
"output_format": "gif",
"frame_count": 9,
"frame_duration_ms": 200
}
Sleepy coffee robot
GIF · 512×512Auto-looping · 9 frames · 200 ms per frame
Prompt & generation settings
Create a looping reaction GIF meme. A sleepy round purple robot holding a coffee mug slowly opens its eyes, lifts the mug a little, and returns to the starting pose. Minimal flat illustration on a pale lavender background, fixed camera. The caption LOADING stays completely still and readable throughout. Smooth seamless loop, consistent character.
{
"n": 1,
"complexity": "low",
"model": "openai/gpt-image-2.6-sunburst",
"prompt": "Create a looping reaction GIF meme. A sleepy round purple robot holding a coffee mug slowly opens its eyes, lifts the mug a little, and returns to the starting pose. Minimal flat illustration on a pale lavender background, fixed camera. The caption LOADING stays completely still and readable throughout. Smooth seamless loop, consistent character.",
"size": "512x512",
"quality": "low",
"output_format": "gif",
"frame_count": 9,
"frame_duration_ms": 200
}
Skeptical SURE cat
GIF · 512×512Auto-looping · 9 frames · 200 ms per frame
Prompt & generation settings
Create a square looping reaction GIF meme. Close-up portrait of an original fluffy round gray cat with expressive amber eyes. It looks straight ahead, slowly narrows one eye and glances sideways with gentle skepticism, then blinks and returns to the initial expression. Tactile soft clay-like 3D illustration, lilac studio background, soft even light, fixed camera, no camera movement. Big dark lettering across the top reads exactly SURE and stays perfectly still and legible in every frame. Tiny controlled facial animation, consistent face, seamless loop. No other text, logos or watermarks.
{
"model": "openai/gpt-image-2.6-sunburst",
"prompt": "Create a square looping reaction GIF meme. Close-up portrait of an original fluffy round gray cat with expressive amber eyes. It looks straight ahead, slowly narrows one eye and glances sideways with gentle skepticism, then blinks and returns to the initial expression. Tactile soft clay-like 3D illustration, lilac studio background, soft even light, fixed camera, no camera movement. Big dark lettering across the top reads exactly SURE and stays perfectly still and legible in every frame. Tiny controlled facial animation, consistent face, seamless loop. No other text, logos or watermarks.",
"size": "512x512",
"quality": "low",
"complexity": "low",
"output_format": "gif",
"n": 1,
"frame_count": 9,
"frame_duration_ms": 200
}
Five-minute fix, three hours later
PNG · 1024×1024Compressed preview; dimensions refer to the original PNG output.
Prompt & generation settings
Create a square 1024x1024 static meme illustration with two equally sized horizontal panels stacked vertically, a clean white gutter between them. An original round raccoon software developer appears in both panels with consistent design. Top panel: relaxed raccoon at a tidy laptop with a small coffee, bright morning light; large caption at the top reads exactly 5 MINUTE FIX. Bottom panel: same raccoon much later with tired wide eyes, surrounded by blank sticky notes and three coffee cups, abstract tangled lines on the screen with no readable code; large caption at the top reads exactly 3 HOURS LATER. Funny expressive hand-drawn comic, bold outlines, restrained teal and orange palette, highly legible black typography on clear pale header bands. Keep both captions spelled exactly. No other text, brands or watermarks.
{
"model": "openai/gpt-image-2.6-sunburst",
"prompt": "Create a square 1024x1024 static meme illustration with two equally sized horizontal panels stacked vertically, a clean white gutter between them. An original round raccoon software developer appears in both panels with consistent design. Top panel: relaxed raccoon at a tidy laptop with a small coffee, bright morning light; large caption at the top reads exactly 5 MINUTE FIX. Bottom panel: same raccoon much later with tired wide eyes, surrounded by blank sticky notes and three coffee cups, abstract tangled lines on the screen with no readable code; large caption at the top reads exactly 3 HOURS LATER. Funny expressive hand-drawn comic, bold outlines, restrained teal and orange palette, highly legible black typography on clear pale header bands. Keep both captions spelled exactly. No other text, brands or watermarks.",
"size": "1024x1024",
"quality": "low",
"complexity": "low",
"output_format": "png",
"n": 1
}
Chinese meme: one small revision
PNG · 1024×1024Compressed preview; dimensions refer to the original PNG output.
Prompt & generation settings
Create a square 1024x1024 static Chinese meme, two equally sized panels side by side with a clean vertical gutter. Original cute capybara designer character, consistent in both panels, warm editorial comic with dark brown outlines and ivory paper background. Left panel: a very small neat blank sticky note and a relaxed capybara; heading at the top in large bold Chinese reads exactly 客户说; speech bubble reads exactly 小改一下. Right panel: the same capybara with a stunned expression behind a comically tall stack of colorful blank design sheets; heading at the top in large bold Chinese reads exactly 我理解的; speech bubble reads exactly 重新做一版. Render all four Chinese phrases accurately, clear and readable, horizontal text, generous spacing. No other text, logos or watermarks.
{
"model": "openai/gpt-image-2.6-sunburst",
"prompt": "Create a square 1024x1024 static Chinese meme, two equally sized panels side by side with a clean vertical gutter. Original cute capybara designer character, consistent in both panels, warm editorial comic with dark brown outlines and ivory paper background. Left panel: a very small neat blank sticky note and a relaxed capybara; heading at the top in large bold Chinese reads exactly 客户说; speech bubble reads exactly 小改一下. Right panel: the same capybara with a stunned expression behind a comically tall stack of colorful blank design sheets; heading at the top in large bold Chinese reads exactly 我理解的; speech bubble reads exactly 重新做一版. Render all four Chinese phrases accurately, clear and readable, horizontal text, generous spacing. No other text, logos or watermarks.",
"size": "1024x1024",
"quality": "low",
"complexity": "low",
"output_format": "png",
"n": 1
}
4K paper-cut mountains and sun
PNG · 3840×2160Compressed preview; dimensions refer to the original PNG output.
Prompt & generation settings
A wide editorial landscape of layered paper-cut mountains and a small warm orange sun, ivory sky, indigo and teal ridgelines, crisp intricate paper edges, balanced negative space. No text.
{
"n": 1,
"complexity": "low",
"model": "openai/gpt-image-2.6-sunburst",
"prompt": "A wide editorial landscape of layered paper-cut mountains and a small warm orange sun, ivory sky, indigo and teal ridgelines, crisp intricate paper edges, balanced negative space. No text.",
"size": "3840x2160",
"quality": "high",
"output_format": "png"
}- Input modalities
- Text
- Output modalities
- Image
- Capabilities
- Image generation
- Upstream release
- 2026-11-04
- RouteMux name
- openai/gpt-image-2.6-sunburst
What is GPT Image 2.6 Sunburst
GPT Image 2.6 Sunburst is one of the two API models OpenAI released with Images 2.6 on November 4, 2026. OpenAI positions it for detailed image generation and editing: at medium or high quality, it says Sunburst produces higher-quality images and keeps characters, products and brand details more consistent across edits. For fast iteration, OpenAI points to GPT Image 2.6 Flare instead. They are separate model ids, so you choose by name.
The headline addition is GIF output. Set output_format to gif and a single request returns a complete looping GIF, with no frame assembly on your side. On RouteMux a GIF is 512×512, frame_count is 2 to 16, frame_duration_ms is 20 to 2000 in steps of 10, and n must be 1. In our runs on November 6, 2026, a 9-frame GIF at low quality used 1,172 image output tokens and took Sunburst 29 to 35 seconds, slower than Flare. You can pick GIF in the try-it panel on this page and run one.
Still images come back as PNG, with size given in pixels such as 1024x1024, 2048x2048 or 3840x2160; unlike 2.5, the literal "2K" and "4K" values are not accepted. Quality runs from auto through low, medium, high and xhigh to max, and complexity (low, medium, high) sets how much planning goes into each image. Billing is per token, with separate rates for text input and image output that stay the same at standard, 2K and 4K, so larger images and higher quality simply produce more output tokens.
Requests use the OpenAI Images protocol at /v1/images/generations and return synchronously, with the image as base64 in data[0].b64_json. OpenAI's 2.6 also supports edits and SVG export; RouteMux has not opened those yet.
GPT Image 2.6 Sunburst API pricing
Prices verified 2026-11-05
Estimate cost
Key capabilities
- GIFs in one request
- Set
output_formatto gif for a complete looping GIF: 512×512, 2 to 16 frames, 20 to 2000 ms per frame. - Detail and consistency
- OpenAI says Sunburst delivers higher quality and keeps characters and products consistent across edits at medium or high quality.
- 2K / 4K stills
- Give size in pixels, up to 3840×2160; standard, 2K and 4K share one set of token rates.
- 20% below official rates
- Billed per token on RouteMux, with text input and image output each 20% below OpenAI list price.
Common use cases
- Reaction GIFs and memes
- The first use OpenAI lists in its launch post: reactions and memes for social media.
- Consistent mascots
- OpenAI mentions animating a brand mascot for a campaign while keeping character, product and brand details consistent.
- Animated storyboards
- Explore character movement with animated storyboards before video production, an example from the launch post.
Limited time10% extra Credits on your first top-up — up to 200 bonus Credits
Applies to your first top-up within 72 hours of signing up. Minimum $20, once per account.
Interface and protocols
Use any of these IDs to call this model via the API.
openai/gpt-image-2.6-sunburstReplace the ROUTEMUX_KEY placeholder with your API key. Create one →
import base64
from pathlib import Path
from openai import OpenAI
client = OpenAI(
base_url="https://api.routemux.com/v1",
api_key="<ROUTEMUX_KEY>",
)
response = client.images.generate(
model="openai/gpt-image-2.6-sunburst",
prompt="A red panda coding under the moonlight",
n=1,
)
Path("image.png").write_bytes(base64.b64decode(response.data[0].b64_json))Performance (last 30 days)
GPT Image 2.6 Sunburst FAQ
›What is GPT Image 2.6 Sunburst?
Generate PNG images and GIF memes, with 2K/4K output and token pricing.
›How much does the GPT Image 2.6 Sunburst API cost?
Through RouteMux, GPT Image 2.6 Sunburst is priced at $24, below the official list price.
›How do I call GPT Image 2.6 Sunburst via API?
GPT Image 2.6 Sunburst is OpenAI-compatible: point your base URL at RouteMux and set the model field to openai/gpt-image-2.6-sunburst — no code changes needed.
›Which input and output modalities does GPT Image 2.6 Sunburst support?
GPT Image 2.6 Sunburst accepts text as input and produces image as output.
›When was GPT Image 2.6 Sunburst released?
GPT Image 2.6 Sunburst was released on November 4, 2026.