GPT Image 2 AI Image Generator
Generate AI images from text prompts or edit existing images. 1K, 2K, or 4K resolution with three quality levels. 11 preset aspect ratios or custom dimensions up to 3840px.
GPT Image 2 AI Image Generator
GPT Image 2 image examples
Sample outputs generated with GPT Image 2.
Landscape Cinematic
A cinematic reference frame for film and video projects.
GPT Image 2 generates and edits AI images at up to 4K with three quality tiers
GPT Image 2 is an image generation and editing model that creates images from text prompts or edits existing images using multi-reference input. It supports 1K, 2K, and 4K resolution with low, medium, and high quality levels. Use 11 preset aspect ratios or define custom dimensions up to 3840px.
Write a prompt up to 20,000 characters describing the image you want. Choose resolution, quality, and aspect ratio.
Upload one or more reference images and describe the edit. GPT Image 2 Edit mode processes multi-image input.
Use preset ratios (1:1, 16:9, 9:16, etc.) or define custom WIDTHxHEIGHT up to 3840px with both dimensions divisible by 16.
When to use GPT Image 2 over other image models
4K native output
Generate at 4K resolution directly for maximum detail. Use for print, large displays, and high-resolution assets.
Three quality levels
Low for fast iteration, medium for balanced output, high for maximum detail. Each level has different pricing.
Multi-image editing
Upload multiple reference images for editing. Combine elements from different sources into one output.
Custom aspect ratios
11 presets plus custom dimensions. Both dimensions divisible by 16, max 3840px, ratio up to 3:1.
GPT Image 2 image generation capabilities
1K, 2K, and 4K resolution
Choose output resolution. 1K for drafts, 2K for production, 4K for maximum detail.
Low, medium, and high quality
Three quality tiers with different cost and detail levels. Low is default.
11 preset aspect ratios
auto, 1:1, 2:3, 3:2, 4:3, 3:4, 4:5, 5:4, 16:9, 9:16, 21:9.
Custom dimensions
Define WIDTHxHEIGHT (e.g., 2304x1536). Both divisible by 16, max 3840px edge, ratio up to 3:1.
20,000-character prompts
Describe the image in detail with up to 20,000 characters for precise control.
Multi-image edit input
Upload multiple reference images for the edit model. Combine and transform visual elements.
GPT Image 2 technical specifications
- Resolution
- 1K, 2K, 4K
- Quality
- Low, Medium, High
- Preset sizes
- auto, 1:1, 2:3, 3:2, 4:3, 3:4, 4:5, 5:4, 16:9, 9:16, 21:9
- Custom size
- WIDTHxHEIGHT (divisible by 16, max 3840px, ratio ≤ 3:1)
- Pixel range
- 655,360 – 8,294,400 pixels
- Prompt length
- Up to 20,000 characters
- Output
- 1 image per request
- Processing
- Asynchronous (task-based)
- Edit input
- Multiple images via image_urls array
What to create with GPT Image 2
Product visuals
Generate product images for e-commerce catalogs, ads, and social media.
Concept art
Create reference frames, mood boards, and visual concepts for video production.
Social media graphics
Generate images at 9:16, 1:1, or 16:9 for Instagram, TikTok, and YouTube thumbnails.
Image editing
Edit existing product photos, swap backgrounds, and refine visual elements.
Generate images with GPT Image 2 in three steps
GPT Image 2 FAQ
Generate images with GPT Image 2
Text-to-image and image editing at 1K, 2K, or 4K with three quality levels. Preview credits before every generation.