Generate
Back to Models
Models/openai/image · text-to-image · image-editingPass-through

GPT Image 2

GPT Image 2 is built for detailed compositions, structured creative briefs, fine typography, and targeted natural-language edits. MachGen supports text-to-image and editing with up to 16 ordered input images across eight exact aspect-and-resolution grids including 21:9, from 1024p through 4096p.

Current MachGen support
Text to ImageImage Editing1K / 1080p / 2K / 4KUp to 16 input images
Starting at $0.048 $0.04 / image with auto-reload
Auto-reloadRegular
1280×544$0.04/image$0.048/image
1280×720$0.057/image$0.068/image
1920×816$0.052/image$0.063/image
1280×848$0.071/image$0.085/image
2048×880$0.056/image$0.067/image
1280×960$0.084/image$0.10/image
+16 more price tiers in the full pricebookLive public pricebookSee full pricebook
Overview

GPT Image 2

GPT Image 2 is OpenAI's latest image model for extremely detailed stills with fine typography and strong instruction following. It is built for briefs that already read like product or design prompts - including precise edits to existing images.

MachGen supports text-to-image and editing with up to 16 ordered input images across eight exact aspect-and-resolution grids including 21:9. Available tiers depend on the selected aspect ratio.

Model guide

Why choose GPT Image 2

Key strengths

  • High-fidelity image generation for detailed briefs, structured compositions, and precise natural-language revisions.
  • Strong typography and instruction following for layouts that combine imagery and copy.
  • Fixed output presets cover common square, portrait, landscape, and high-resolution delivery formats.

Good fit for

  • Product imagery, posters, packaging concepts, UI illustrations, and campaign stills.
  • Fine-grained edits to an existing image without rebuilding the entire composition.
  • Creative briefs containing explicit layout, material, lighting, and copy requirements.
Supported on MachGen

Inputs and output settings

This table reflects the model routes and controls currently exposed by MachGen.

ModeRequired inputAvailable settings
Text to ImageText promptExact output grid: 1:1: 1024p 1280x1280, 2048p 2048x2048; 16:9: 1024p 1280x720, 1080p 1920x1088, 2048p 2048x1152, 4096p 3840x2160; 21:9: 1024p 1280x544, 1080p 1920x816, 2048p 2048x880, 4096p 3840x1648; 9:16: 1024p 720x1280, 1080p 1088x1920, 2048p 1152x2048, 4096p 2160x3840; 4:3: 1024p 1280x960, 2048p 2048x1536; 3:4: 1024p 960x1280, 2048p 1536x2048; 3:2: 1024p 1280x848, 2048p 2048x1360; 2:3: 1024p 848x1280, 2048p 1360x2048
Image EditingText prompt and up to 16 source imagesExact output grid: 1:1: 1024p 1280x1280, 2048p 2048x2048; 16:9: 1024p 1280x720, 1080p 1920x1088, 2048p 2048x1152, 4096p 3840x2160; 21:9: 1024p 1280x544, 1080p 1920x816, 2048p 2048x880, 4096p 3840x1648; 9:16: 1024p 720x1280, 1080p 1088x1920, 2048p 1152x2048, 4096p 2160x3840; 4:3: 1024p 1280x960, 2048p 2048x1536; 3:4: 1024p 960x1280, 2048p 1536x2048; 3:2: 1024p 1280x848, 2048p 2048x1360; 2:3: 1024p 848x1280, 2048p 1360x2048
Workflow

Using GPT Image 2

  1. Choose a supported generation mode in Playground.
  2. Add the required prompt and source or reference assets.
  3. Select output settings from the controls shown for that mode.
  4. Generate, then inspect the returned asset before reusing the settings through the API.

Practical guidance

  • Describe the subject, action or composition, environment, and visual direction in a clear order.
  • For image editing, state what should change and what should remain unchanged.
Capabilities

What you can make

Text to image

Generate richly detailed compositions with accurate small text and structured layouts from natural language.

Image editing

Make fine-grained revisions from up to 16 ordered input images - local fixes, style shifts, or object changes - while keeping the rest of the frame coherent.

Output controls

Select one of eight aspect ratios, including 21:9, and an exact resolution tier available for that ratio.

Prompting

Prompting notes

  • Put exact strings in quotes and specify placement ("title centered at top in bold sans").
  • For edits, describe the before/after clearly rather than only the desired end state.
  • Choose one of the exact resolution tiers available for the selected aspect ratio.
Best results

How to get better GPT Image 2 results

  • Describe the scene as a design brief with clear priorities instead of a long stream of adjectives.
  • Put exact copy in quotation marks and specify placement, alignment, and hierarchy.
  • For editing, explain both the desired change and the elements that must be preserved.
  • Inspect hands, small text, logos, and repeated patterns closely before using the image as a final asset.
Limits

Limits and availability

  • MachGen admits only the modes and settings listed above for this catalog model.
  • Source assets must finish uploading before a request can be submitted.
  • Generated results can vary between requests, including when the same prompt and settings are reused.
  • GPT Image 2 is pass-through. MachGen sends the request through its upstream provider integration and returns the generated asset.
Learn more

Learn more

For model background, technical details, and architecture, see the official OpenAI image generation guide.

Try this model in the MachGen Playground.

Official documentation may describe capabilities or parameters that are not currently exposed by MachGen. Use the Inputs and output settings section above as the source of truth for this page.

Pricing

GPT Image 2 pricing

See Pricing for currently published MachGen rates. When available, the Playground estimate reflects the selected task and output settings.

On MachGen

GPT Image 2 on MachGen

Deployment: Pass-through. GPT Image 2 is pass-through. MachGen sends the request through its upstream provider integration and returns the generated asset.

The Playground and API sections on this page list the inputs and output settings exposed for this model.