GPT-image-2
Version: 2026-04-21
Direct from Azure models
Direct from Azure models are a select portfolio curated for their market-differentiated capabilities:- Secure and managed by Microsoft: Purchase and manage models directly through Azure with a single license, consistent support, and no third-party dependencies, backed by Azure's enterprise-grade infrastructure.
- Streamlined operations: Benefit from unified billing, governance, and seamless PTU portability across models hosted on Azure - all part of Microsoft Foundry.
- Future-ready flexibility: Access the latest models as they become available, and easily test, deploy, or switch between them within Microsoft Foundry; reducing integration effort.
- Cost control and optimization: Scale on demand with pay-as-you-go flexibility or reserve PTUs for predictable performance and savings.
Key capabilities
About this model
GPT-image-2 is OpenAI's latest cutting-edge image generation model. Key value adds include better performance, quality, editing controls, and face preservation.The model supports high input_fidelity and adding/removing one aspect of the image while retaining others. This model includes improvements in aspect ratio, resolution, and editing capabilities.
Key model capabilities
- Includes all capabilities of gpt-image-1, gpt-image-2
- Text to image generation
- Image to image generation
- Inpainting
- High quality image generations: supports any resolution in size parameter, as long as it meets the following constraints:
- The maximum edge length must be less than 3840px
- Both edges must be a multiple of 16
- The ratio between the long edge and the short edge must not be greater than 3:1
- The total maximum number of pixels in the requested resolution must not exceed 8,294,400.
- The total minimum number of pixels in the requested resolution must not be less than 655,360.
- Face preservation
Use cases
Retail: Generate product imagery for catalogs, e-commerce listings, and personalized shopping experiences. Marketing: Create campaign visuals, social media graphics, and targeted advertisements at scale.Out of scope use cases
The provider has not supplied this information.Pricing
Pricing is based on a number of factors, including deployment type and tokens used. See pricing details here.Key Documentation
The following documents are applicable:- https://learn.microsoft.com/en-us/azure/ai-foundry/openai/how-to/dall-e?tabs=gpt-image-1
- https://learn.microsoft.com/en-us/azure/ai-foundry/openai/latest
- Overview of Responsible AI practices for Azure OpenAI models
- Transparency Note for Azure OpenAI Service
- Policies intended to block prohibited IP production
- Block reproduction of photorealistic persons when they appear on blocklist
- Block image input with photorealistic depictions of minors.
- All image output will attach C2PA data
- Microsoft's abuse monitoring and content safety will prevent disallowed content.
Technical specs
The provider has not supplied this information.Training cut-off date
The provider has not supplied this information.Training time
The provider has not supplied this information.Input formats
The provider has not supplied this information.Output formats
High quality image generations with flexible aspect ratios and resolutionSupported languages
The provider has not supplied this information.Sample JSON response
The provider has not supplied this information.Model architecture
The provider has not supplied this information.Long context
The provider has not supplied this information.Optimizing model performance
The provider has not supplied this information.Additional assets
The provider has not supplied this information.Training disclosure
Training, testing and validation
The provider has not supplied this information.Distribution
Distribution channels
This model is provided through the Azure OpenAI Service.Responsible AI considerations
Safety techniques
Despite technical mitigations such as data filtering, image generation models are known to produce harmful or unexpected content based on user requests. In addition to technical work on the model such as data filtering, additional mitigations are also applied at the system level to further enhance end user safety (e.g., content classifiers). Some common risk areas associated with image generation models include violent or gory content, sexual content or nudity, depictions of public figures, replication of trademarked or other protected material.Safety evaluations
The provider has not supplied this information.Known limitations
The provider has not supplied this information.Acceptable use
Acceptable use policy
The provider has not supplied this information.Model Specifications
LicenseCustom
Last UpdatedApril 2026
Input TypeText,Image
Output TypeImage
ProviderOpenAI
Languages1 Language