ホーム
アセット
マイ作品
Models
ボイスライブラリ
開発者
ブログ
日本語
料金
ログイン
Models/Image/GPT-Image-2

OpenAI: GPT-Image-2

  • Text to image
  • Image to image
  • Multi-reference
API referenceUse in AI画像生成エディタ

The strongest instruction following of the image models — it understands edits and local replacements.

It also lets you omit `aspectRatio` and have the model choose, which none of the others do.

Strengths

  • Strong instruction following for edits and local replacements
  • `aspectRatio` can be omitted and left to the model
  • Eight standard ratios

Limitations

  • Four reference images, the lowest in the image line
  • No 1:4, 4:1, 1:8 or 8:1

Capabilities

CapabilityValue
Text to image
Yes
Image to imageYes
Multi-referenceYes
Precise edit–No
Quality tiers–No

Specs and limits

ParameterValue
Image size1K / 2K / 4K
Aspect ratio1:1 · 2:3 · 3:2 · 3:4 · 4:3 · 4:5 · 5:4 · 9:16 · 16:9 · 21:9
Reference images4
Quality tiersNo

How it is billed

按张计费，随尺寸档变化

Standard pricesCredits
1K4
2K6
4K10

Figures come from the same estimator the generation form calls before you spend anything.

Alternatives

  • Seedream 5.0 ProWhen you need up to 10 reference images, or precise editing.
  • Nano Banana ProWhen there is text in the frame, or several references have to agree.

Call it

# Set LISTENHUB_API_KEY firstcurl -X POST "https://api.marswave.ai/openapi/v1/images/generation" \-H "Authorization: Bearer $LISTENHUB_API_KEY" \-H "Content-Type: application/json" \-d '{"provider": "google","model": "gpt-image-2","prompt": "A serene mountain landscape at sunset with a reflective lake","imageConfig": {"aspectRatio": "16:9"}}'
  • API reference
  • SDK & CLI
  • Agent Skills

Similar models

  • Nano Banana Pro
  • Nano Banana Flash
  • GPT-Image-2.5 Flare
  • GPT-Image-2.5 Sunburst
  • GPT-Image-2 Pro