logo
RunComfy
  • ComfyUI
  • TrainerNew
  • Models
  • API
  • Pricing
discord logo
MODELS
Explore
All Models
LIBRARY
Generations
MODEL APIS
API Docs
API Keys
ACCOUNT
Usage

GPT Image 2.6 Sunburst Image Edit: Precise Image Editing on Models and API | RunComfy

openai/gpt-image-2.6/sunburst/edit

Upload pictures, describe the change, and GPT Image 2.6 Sunburst Image Edit returns a detailed edit that keeps untouched areas consistent.

Image 1
Reference images to edit or combine, up to 16. Mask and markup apply to the first image.
Optional mask for the first image: white marks the area to edit, black the area to keep. Use either a mask or a markup image.
Optional transparent PNG with circles, arrows, or notes drawn over the first image. Use either a mask or a markup image.
Shape of the edited image. auto follows the first input image.
Output resolution tier.
Rendering quality. Higher tiers add detail and cost more.
How much the model plans the layout before drawing. Does not change the price.
File format of the edited image.
Idle
Price per output image by quality x resolution, including one input image (1K / 2K / 4K): low $0.064 / $0.095 / $0.159, medium $0.095 / $0.154 / $0.254, high $0.111 / $0.175 / $0.301, xhigh $0.185 / $0.301 / $0.523, max $0.396 / $0.650 / $1.072. Each additional reference image beyond the first adds $0.016, and a mask or markup image adds $0.016.

Introduction To GPT Image 2.6 Sunburst Image Edit

OpenAI's GPT Image 2.6 Sunburst Edit applies written instructions to existing pictures with careful detail, taking up to 16 references plus an optional mask or markup, at up to 4K.
Trading hours of retouching for exact, repeatable changes, GPT Image 2.6 Sunburst Image Edit fits brand, retail, and agency teams that need polished results.
For developers, GPT Image 2.6 Sunburst Image Edit on RunComfy can be used both in the browser and via an HTTP API, so you don't need to host or scale the model yourself.
Ideal for: Campaign Asset Retouching | Product Image Localization | Multi-Step Brand Edits

OpenAI / GPT Image 2.6 Sunburst Edit#


GPT Image 2.6 Sunburst Image Edit is the precision editing mode of the GPT Image 2.6 family. It takes longer per image than Flare, and uses that time to keep faces, products, and textures stable while it applies your changes.


GPT Image 2.6 Sunburst Image Edit holds detail well across repeated passes, so you can chain several edits on one asset without it drifting away from the original.


Highlights#


  • Steady detail: GPT Image 2.6 Sunburst Image Edit keeps untouched regions consistent, even after several rounds of changes.

  • Up to 16 references: GPT Image 2.6 Sunburst Image Edit can merge products, people, and backgrounds from several pictures into one scene.

  • Exact targeting: In GPT Image 2.6 Sunburst Image Edit, a mask limits which pixels may change; a markup image points to the spot with circles, arrows, or notes.

  • Two separate dials: Quality sets detail and price, while complexity only sets how much layout planning happens.

  • Shape control: GPT Image 2.6 Sunburst Image Edit keeps the first image's ratio with auto, or you can pick one of fifteen fixed ratios, at 1K, 2K, or 4K.

Parameters#


ParameterRequiredTypeDefaultRange / OptionsDescription
images*Yes (*)Array of URLs—1 – 16 imagesSource pictures to edit or combine; the first one is the main canvas.
prompt*Yes (*)String——Instruction describing what to change and what to keep.
mask_imageNoImage URL——White areas get edited, black areas stay untouched (first image only).
markup_imageNoImage URL——Transparent PNG with circles, arrows, or notes on top of the first image.
aspect_ratioNoStringautoauto, 1:1, 3:2, 2:3, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9, 2:1, 1:2, 3:1, 1:3, 9:21auto keeps the shape of the first image.
resolutionNoString1k1k, 2k, 4kOutput size tier.
qualityNoStringmediumlow, medium, high, xhigh, maxRendering effort; higher tiers cost more.
complexityNoStringhighlow, medium, highLayout planning before drawing; same price at every level.
output_formatNoStringpngpng, jpeg, webpFile type of the edited image.

Pricing#


Quality1K2K4K
low$0.064$0.095$0.159
medium$0.095$0.154$0.254
high$0.111$0.175$0.301
xhigh$0.185$0.301$0.523
max$0.396$0.650$1.072

Prices are per output image and include one input image.


  • Extra reference images: each image after the first adds $0.016.
  • Mask or markup image: adds $0.016.
  • Complexity never changes the price.
  • Example: medium quality at 1K with three reference images costs $0.095 + 2 × $0.016 = $0.127.

How to Use#


  1. Open GPT Image 2.6 Sunburst Image Edit on its RunComfy model page.
  2. Upload the main picture first, then any extra references.
  3. Write the instruction for GPT Image 2.6 Sunburst Image Edit, listing both the change and the parts that must not move.
  4. Add a mask or a markup image for the first picture if the edit needs an exact location.
  5. Keep aspect ratio on auto or choose a new frame.
  6. Choose resolution and quality; high or xhigh suits final deliverables.
  7. Generate with GPT Image 2.6 Sunburst Image Edit, inspect fine details, and run a follow-up edit on the result if needed.

Prompt & Reference Tips#


  • Protect key areas: State plainly which faces, logos, or products must remain identical, and GPT Image 2.6 Sunburst Image Edit will leave them alone.

  • Chain edits: GPT Image 2.6 Sunburst Image Edit holds up well over several passes, so split large changes into steps.

  • Order references: GPT Image 2.6 Sunburst Image Edit treats the first image as the canvas for masks, markup, and auto ratio.

  • Point with markup: A quick arrow or circle often beats a long written description of a location.

  • Quote new text: For GPT Image 2.6 Sunburst Image Edit, put added words in quotes and describe font and placement.

  • Match lighting: Mention light direction when inserting objects so GPT Image 2.6 Sunburst Image Edit blends the shadows in.

How GPT Image 2.6 Sunburst Edit compares to other models#


  • Versus GPT Image 2.6 Flare Image Edit: Flare returns edits faster, while GPT Image 2.6 Sunburst Image Edit keeps finer detail stable across passes.

  • Versus GPT Image 2.5 Sunburst edit: The 2.6 tier adds mask and markup inputs plus the complexity control, based on publicly available information.

  • Versus classic inpainting tools: GPT Image 2.6 Sunburst Image Edit works from full sentences and multiple references, reducing manual selection work.

Related Models

nano-banana/edit

Edit and blend images with prompts using Google Nano Banana.

nano-banana-2/text-to-image

Fast, high-quality text-to-image generation with Nano Banana 2, with aspect ratio and resolution controls.

flux-2/flex/edit

High-accuracy image transformation model with color control and creative precision for visual professionals.

qwen-edit-2509/lora/fusion

Blend and refine visuals with advanced image editing, depth control, and multilingual design precision.

flux-2/lora

Advanced text-to-image system with LoRA adapters, style control, and photoreal accuracy for design professionals.

flux-2/pro/text-to-image

Create reliable, studio-grade visuals with precise color and layout control.

Frequently Asked Questions

What is GPT Image 2.6 Sunburst Image Edit best used for?

GPT Image 2.6 Sunburst Image Edit applies written instructions to existing pictures while keeping faces, products, and textures stable. Brand, retail, and agency teams use GPT Image 2.6 Sunburst Image Edit for campaign retouching, product localization, and multi-step edits on hero assets.

How is GPT Image 2.6 Sunburst Image Edit different from the Flare edit model?

GPT Image 2.6 Sunburst Image Edit spends more time per image to hold fine detail steady, especially across repeated passes on the same asset. The Flare edit model returns results faster, so it is often used for exploration before the final version is made with GPT Image 2.6 Sunburst Image Edit.

Can GPT Image 2.6 Sunburst Image Edit combine several reference images?

Yes. GPT Image 2.6 Sunburst Image Edit accepts up to 16 input images, letting you merge products, people, and backgrounds into one scene. The first image is the canvas, so GPT Image 2.6 Sunburst Image Edit applies masks, markup, and auto aspect ratio to it.

How do I target a specific area with GPT Image 2.6 Sunburst Image Edit?

Add a mask_image where white marks the area GPT Image 2.6 Sunburst Image Edit may change and black marks the area to keep. Alternatively, upload a markup_image with circles, arrows, or notes over the first image; GPT Image 2.6 Sunburst Image Edit accepts one or the other, not both.

Can I chain multiple edits with GPT Image 2.6 Sunburst Image Edit?

Yes, GPT Image 2.6 Sunburst Image Edit is designed to keep untouched regions consistent across repeated edits. Splitting a large change into several smaller passes on GPT Image 2.6 Sunburst Image Edit usually gives more predictable results than one long instruction.

What output options does GPT Image 2.6 Sunburst Image Edit offer?

GPT Image 2.6 Sunburst Image Edit outputs png, jpeg, or webp at 1k, 2k, or 4k, with quality from low to max and complexity from low to high. Aspect ratio can stay on auto to follow the first image or use one of fifteen fixed ratios; check the RunComfy parameter panel for GPT Image 2.6 Sunburst Image Edit for current limits.

Can developers access GPT Image 2.6 Sunburst Image Edit via the RunComfy API?

Yes. Prototype your edits with GPT Image 2.6 Sunburst Image Edit in the RunComfy model UI, then send identical parameters through the RunComfy API in production. GPT Image 2.6 Sunburst Image Edit runs as a hosted endpoint, so there is no infrastructure for you to manage.

How much does GPT Image 2.6 Sunburst Image Edit cost on RunComfy?

GPT Image 2.6 Sunburst Image Edit is billed per output image by quality and resolution, including one input image, from $0.064 for low quality at 1K to $1.072 for max quality at 4K; medium at 1K costs $0.095. Each extra reference image adds $0.016, and a mask or markup image adds $0.016 with GPT Image 2.6 Sunburst Image Edit.

Follow us
  • LinkedIn
  • Facebook
  • Instagram
  • Twitter
Support
  • Discord
  • Email
  • System Status
  • Affiliate
Video Models
  • Kling 4.0 Image to Video
  • Kling 4.0 Reference to Video
  • Kling 4.0 Video Editing
  • Wan 3.0
  • Wan 2.5
  • MiniMax H3 Fast
  • View All Models →
Image Models
  • Seedream 4.5
  • Seedream 5.0 Pro
  • Flux 2 Pro
  • Qwen Image Edit 2511 LoRA
  • seedream 4.0
  • Nano Banana Pro
  • View All Models →
Legal
  • Terms of Service
  • Privacy Policy
  • Cookie Policy
RunComfy
Copyright 2026 RunComfy. All Rights Reserved.

RunComfy is the premier ComfyUI platform, offering ComfyUI online environment and services, along with ComfyUI workflows featuring stunning visuals. RunComfy also provides AI Models, enabling artists to harness the latest AI tools to create incredible art.

Examples Of GPT Image 2.6 Sunburst Image Edit