GPT Image 2.0, OpenAI's latest image model, is capable of making fine-grained, detailed edits to images.
High quality (default)
Output
Example
Example output from GPT Image 2 Edit
Pricing
Criteria
High quality (default)
per output image
or around 75 images for $10
Overview
GPT Image 2 Edit by OpenAI applies fine-grained, detail-preserving edits to existing images from a natural-language prompt. Best-in-class for typography edits, compositional changes, and reference-guided transformations.
Key capabilities
- ●$0.009–$0.133 per image (low/medium/high quality) — usage-based.
- ●Multi-region failover, Runflow runs optimized inference across multiple regions. If one region has an outage, traffic routes to another automatically.
- ●Multi-image reference, provide up to several reference images for composition-aware editing.
- ●Optional mask, target a specific region of the image with `mask_url` for precise inpainting.
GPT Image family
| Model | Details |
|---|---|
| GPT Image 2 Edit (this model) | Image editing, latest generation |
| GPT Image 2 | Text-to-image sibling |
| GPT Image 1.5 Edit | Previous generation editor |
| GPT Image 1.5 | Previous generation text-to-image |
Technical specifications
| Spec | Details |
|---|---|
| Quality tiers | low, medium, high (default) |
| Output formats | png, jpeg, webp |
| Inputs | prompt, image_urls (required), optional mask_url |
| Pricing | $0.009 (low), $0.034 (medium), $0.133 (high) per image |
Frequently asked questions
Related models
GPT Image 1.5
openai/gpt-image-1.5
GPT Image 1.5 generates high-fidelity images with strong prompt adherence, prese...
GPT Image 2
openai/gpt-image-2
GPT Image 2.0, OpenAI's latest image model, is capable of creating extremely det...
Tag Removal
runflow/tag-removal
Remove price tags, stickers, and labels from product images while preserving the...
Product Isolation
runflow/product-isolation
Isolate the main product from a busy scene onto a clean background, using an opt...
Start generating with GPT Image 2 Edit
Get API access in minutes. No GPU setup, no infrastructure to manage.