Public Model / gpt-image

GPT Image 2 API pricing

Prompted image creation and editing with broad aspect-ratio and resolution controls.

Availability
Available
Tasks
Create Image, Edit Image
Published rates from
$0.03

Request calculator

Configure GPT Image 2

Controls and rates come directly from the checked public catalog.

Interactive calculator

Price a direct generation

Choose a task, configure its inputs, and see the current catalog price.

Step 1

Choose a generation type

Step 2

Configure generation

Model
GPT Image 2
Aspect ratioauto
Resolution1K
Outputs1

Step 3

Reference total

$0.031 output at $0.03 each
Generation
$0.031 image x $0.03
Minimum request
$0.03
Get started

The API estimate for your exact request is authoritative before paid work starts.

Best fit

Use GPT Image 2 when a workflow needs both net-new images and edits under one model contract, including high-resolution outputs.

Use cases

  • Campaign key art
  • Reference-guided product edits

Know before building

  • Auto, 3:1, 1:3, 9:21, 5:4 and 4:5 require 1K; square supports 1K or 2K. Background control is available at 1K. Invalid combinations are rejected.

Generated contract

Tasks, controls, and rates

Published rates

image.create
Varies by resolution
$0.03 / image$0.03 minimum
image.edit
Varies by resolution
$0.03 / image$0.03 minimum

Request controls

image.createAspect ratioauto · 1:1 · 3:2 · 2:3 · 4:3 · 3:4 · 16:9 · 9:16 · 2:1 · 1:2 · 3:1 · 1:3 · 21:9 · 9:21 · 5:4 · 4:5
image.createBackgroundtransparent · opaque · auto · null
image.createPromptText input
image.createReference imagesUp to 16 items
image.createResolution1K · 2K · 4K
image.editAspect ratioauto · 1:1 · 3:2 · 2:3 · 4:3 · 3:4 · 16:9 · 9:16 · 2:1 · 1:2 · 3:1 · 1:3 · 21:9 · 9:21 · 5:4 · 4:5
image.editBackgroundtransparent · opaque · auto · null
image.editImagesUp to 16 items
image.editPromptText input
image.editResolution1K · 2K · 4K

Lifecycle: queued -> running -> succeeded -> failed -> cancelled -> manual_review

Output fields: assets

Runnable examples

Estimate first, then create

Both requests use the same normalized task, model, and input payload. Creating work requires an API key and idempotency key.

1. Estimate

curl https://api.entirefeed.com/v1/estimate \
  -X POST \
  -H "Content-Type: application/json" \
  -d '{
  "task": "image.create",
  "model": "gpt-image-2",
  "input": {
    "aspect_ratio": "auto",
    "resolution": "1K",
    "prompt": "A polished campaign still for a minimal running shoe on white"
  }
}'

2. Create

curl https://api.entirefeed.com/v1/generations \
  -X POST \
  -H "Authorization: Bearer $ENTIREFEED_API_KEY" \
  -H "Idempotency-Key: gpt-image-2-example-001" \
  -H "Content-Type: application/json" \
  -d '{
  "task": "image.create",
  "model": "gpt-image-2",
  "input": {
    "aspect_ratio": "auto",
    "resolution": "1K",
    "prompt": "A polished campaign still for a minimal running shoe on white"
  }
}'