HiNow LogoHiNow

GPT Image 2

Model Information

Display Name: GPT Image 2

API Model ID: openai/gpt-image-2

Category: Text To Image

Description: GPT Image 2 is OpenAI's most advanced image generation model, the successor to GPT Image 1.5. It is the first image model with native visual reasoning ("thinking") built into its architecture, enabling state-of-the-art photorealism, superior text rendering, and precise prompt adherence. **Key Features:** - Native visual reasoning for complex multi-element compositions - Accurate multilingual text rendering — menus, infographics, UI mockups, posters - Multi-panel generation with consistent characters, object placement, and brand colors from a single prompt - Flexible resolutions: 1K (fast drafts), 2K (production), 4K (maximum detail, experimental) - Up to 10 images per request **Quality Tiers:** - **Low** — Fast drafts and thumbnails, lowest cost - **Medium** — Balanced quality and speed, recommended default - **High** — Maximum rendering fidelity, ideal for final production assets **Resolutions:** - **1K** (~1280px long edge) — Fast iteration and prototyping - **2K** (~2048px long edge) — Recommended for production use - **4K** (~3840px long edge) — Ultra-high detail (experimental) **Supported Aspect Ratios:** 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3 **Output Formats:** PNG, JPEG, WebP **Best For:** - Professional marketing and advertising visuals - Typography-heavy designs (posters, menus, banners, infographics) - Multi-panel storyboards, comics, and visual narratives - Product visualization and mockups - High-fidelity photorealistic imagery - Multilingual content creation

How to Use This Model

To use GPT Image 2 via the HInow.ai API, use the model ID: openai/gpt-image-2

API Request Example (Chat/Text)


POST https://api.hinow.ai/v1/chat/completions
Authorization: Bearer YOUR_API_KEY
Content-Type: application/json

{
  "model": "openai/gpt-image-2",
  "messages": [
    {"role": "user", "content": "Your message here"}
  ]
}

API Request Example (Image Generation)


POST https://api.hinow.ai/v1/images
Authorization: Bearer YOUR_API_KEY
Content-Type: application/json

{
  "model": "openai/gpt-image-2",
  "prompt": "Your image description here"
}

Pricing

  • total: $0.087

Available Parameters

  • quality: Image quality level. Low is fast and cheap for drafts; Medium balances quality and cost; High delivers maximum fidelity. (Options: low, medium, high)
  • aspect_ratio: Output aspect ratio. Mapped to the closest native OpenAI resolution. (Options: 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3)
  • resolution: Output resolution tier. 1K (~1280px), 2K (~2048px), 4K (~3840px experimental). (Options: 1k, 2k, 4k)
  • output_format: Output image format. (Options: png, jpeg, webp)
  • num_images: Number of images to generate per request. (Options: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10)

Quick Reference

To use this model, set: "model": "openai/gpt-image-2"

Featured: Yes

Documentation: https://hinow.ai/models/openai/gpt-image-2

API Endpoint: https://api.hinow.ai/v1

Back to Models

GPT Image 2

Featured

openai/gpt-image-2

$0.087
per image

About

GPT Image 2 is OpenAI's most advanced image generation model, the successor to GPT Image 1.5. It is the first image model with native visual reasoning ("thinking") built into its architecture, enabling state-of-the-art photorealism, superior text rendering, and precise prompt adherence.

Key Features:

  • Native visual reasoning for complex multi-element compositions
  • Accurate multilingual text rendering — menus, infographics, UI mockups, posters
  • Multi-panel generation with consistent characters, object placement, and brand colors from a single prompt
  • Flexible resolutions: 1K (fast drafts), 2K (production), 4K (maximum detail, experimental)
  • Up to 10 images per request

Quality Tiers:

  • Low — Fast drafts and thumbnails, lowest cost
  • Medium — Balanced quality and speed, recommended default
  • High — Maximum rendering fidelity, ideal for final production assets

Resolutions:

  • 1K (~1280px long edge) — Fast iteration and prototyping
  • 2K (~2048px long edge) — Recommended for production use
  • 4K (~3840px long edge) — Ultra-high detail (experimental)

Supported Aspect Ratios: 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3

Output Formats: PNG, JPEG, WebP

Best For:

  • Professional marketing and advertising visuals
  • Typography-heavy designs (posters, menus, banners, infographics)
  • Multi-panel storyboards, comics, and visual narratives
  • Product visualization and mockups
  • High-fidelity photorealistic imagery
  • Multilingual content creation

Capabilities

Text To ImageImage To Image

Parameters

quality

Image quality level. Low is fast and cheap for drafts; Medium balances quality and cost; High delivers maximum fidelity.

lowmediumhigh
aspect_ratio

Output aspect ratio. Mapped to the closest native OpenAI resolution.

1:116:99:164:33:43:22:3
resolution

Output resolution tier. 1K (~1280px), 2K (~2048px), 4K (~3840px experimental).

1k2k4k
output_format

Output image format.

pngjpegwebp
num_images

Number of images to generate per request.

12345678910

Code Examples

curl -X POST https://api.hinow.ai/v1/images \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $HINOW_API_KEY" \
  -d '{
    "model": "openai/gpt-image-2",
    "prompt": "Transform this image into cyberpunk style",
    "images": [
      "https://example.com/image1.jpg",
      "data:image/png;base64,iVBORw0KGgo..."
    ],
    "parameters": {
      "quality": "low",
      "aspect_ratio": "1:1",
      "resolution": "1k",
      "output_format": "png",
      "num_images": "1"
    }
  }'