Skip to content
CallMissed

Explore CallMissed

Back to all models
Image Generationnewflagship
OpenAI — gpt-image-2.5-sunburst logo

GPT Image 2.5 Sunburst

by OpenAI · Released 2026

OpenAI's most capable image generation and editing model. Built for editing precision — inpainting and reference-image edits that preserve the original. Model id `gpt-image-2.5-sunburst`.

OpenAI — gpt-image-2.5-sunburst logo
Image Generation

GPT Image 2.5 Sunburst

Powered by OpenAI · Proprietary text-to-image with editing

Context Window

N/A

Parameters

Undisclosed

Max Output

N/A

Category

Image Generation

Overview

GPT Image 2.5 Sunburst is OpenAI's most capable model for image generation and editing. Where earlier gpt-image releases were generation-first with editing bolted on, Sunburst is tuned for workflows where the edit itself is the product — inpainting a region without disturbing the rest of the frame, or rebuilding a scene around a logo, product shot or face you supply as a reference.

On CallMissed, call `"model": "gpt-image-2.5-sunburst"` on `/v1/images/generations`, or pass `reference_images` to drive the `/v1/images/edits` surface. It accepts the standard 1024×1024, 1024×1536 and 1536×1024 presets plus custom sizes on a 16-pixel grid up to 3840px per edge, at aspect ratios from 1:3 to 3:1. The quality ladder runs low → medium → high → xhigh → max, with `auto` picking a tier from the prompt.

Priced at $0.25 per image through CallMissed — the same flat rate as GPT Image 2, which runs an identical upstream token meter — so the most capable tier costs no premium over the previous flagship.

Pricing

MetricPrice
Price /image₹25.0000

1 credit = ₹1 = $0.01 USD. Transparent per-call pricing — you pay only for what you call, with no seat fees or minimums.

Key Highlights

  • Most capable OpenAI image generation + editing model
  • Inpainting and reference-image edits
  • Quality tiers up to max
  • Custom sizes to 3840px per edge

Technical Details

  • Model id: gpt-image-2.5-sunburst
  • OpenAI-compatible /v1/images/generations and /v1/images/edits endpoints
  • Input modalities: text, image — output: image
  • Quality: low, medium, high, xhigh, max, auto

Strengths

  • Best-in-class editing precision and inpainting
  • Preserves logos, faces and products from reference images

Limitations

  • Premium per-image cost
  • Slower than Flare for everyday generation

Use Cases

Reference-based brand imageryInpainting and retouchingProduct photographyMarketing visuals

API Example

curl https://api.callmissed.com/v1/images/generations \
  -H "Authorization: Bearer cm_YOUR_KEY" \
  -d '{"model": "gpt-image-2.5-sunburst", "prompt": "A product photo of a ceramic mug on oak, soft window light", "n": 1, "quality": "high"}'

Endpoint: POST /v1/images/generations · Model ID: gpt-image-2.5-sunburst

Try GPT Image 2.5 Sunburst now

Get 1000 free API credits on signup. No credit card required.

Start free Book a DemoRead docs