GPT Image 2 API

Use GPT Image 2 on EvoLink for text-to-image, image editing, and async image delivery with one API key, pricing visibility, and playground access.

Model Type:

Price: $0.015 - 0.039(~ 1.02 - 2.652 credits) per image

Per-call billing tiered by resolution.

May experience occasional downtime. Suitable for development and testing.

Use the same API endpoint. The model parameter selects the variant.

GPT Image 2 API Pricing and Access

Run GPT Image 2 on EvoLink for text-to-image, image-to-image, image editing, and async task delivery. Compare pricing, test prompts in the playground, and move into production with one API.

Prompt

Need inspiration? Browse GPT Image 2 prompts

42 (suggested: 2,000)

Input Images

Upload up to 16 images (max 50MB each)

Click to upload or drag and drop

Supported formats: JPEG, JPG, PNG, WEBP
Maximum file size: 50MB; Maximum files: 16

Size

Ratio mode: w×h ≈ Resolution tier pixel budget. Auto lets the model decide (billed at the 1K tier). Custom takes any WxH — both multiples of 16, 0.65MP–8.29MP total pixels, edges ≤ 3840px, aspect ≤ 3:1. Backend infers the billing tier from total pixels in Custom mode.

Resolution

Resolution tier — drives both the final pixel size and per-call price. Auto and Custom modes ignore this field (auto billed at 1K; Custom mode infers tier from pixels).

Click Generate to see preview

History

Max 20 items

0 running · 0 completed

Your generation history will appear here

Billing Rules

•Per-call billing — one charge per generated image.
•Ratio + resolution: charged by the selected resolution tier.
•`size=auto`: billed at the 1K tier (model decides the output size).
•Explicit `WxH` pixels: tier inferred from total pixels — ≤ 1.7 MP → 1K, > 1.7 MP and ≤ 4.4 MP → 2K, > 4.4 MP → 4K (max 8.29 MP).

Pricing

Model	Mode	Resolution	Price
GPT Image 2 Beta	Image Generation (All Aspect Ratios)	1K (~1.05 MP)	$0.015/ image(1.02 Credits)
GPT Image 2 Beta	Image Generation (All Aspect Ratios)	2K (~4.19 MP)	$0.025/ image(1.734 Credits)
GPT Image 2 Beta	Image Generation (All Aspect Ratios)	4K (~8.29 MP)	$0.039/ image(2.652 Credits)

GPT Image 2 Beta

Image Generation (All Aspect Ratios)

Resolution:1K (~1.05 MP)

Price:

$0.015/ image

(1.02 Credits)

GPT Image 2 Beta

Image Generation (All Aspect Ratios)

Resolution:2K (~4.19 MP)

Price:

$0.025/ image

(1.734 Credits)

GPT Image 2 Beta

Image Generation (All Aspect Ratios)

Resolution:4K (~8.29 MP)

Price:

$0.039/ image

(2.652 Credits)

If it's down, we automatically use the next cheapest available—ensuring 99.9% uptime at the best possible price.

What You Can Build with GPT Image 2

Product Image Refresh Workflows

Use GPT Image 2 to update catalog and marketplace visuals without rebuilding every asset from scratch. Teams can generate new backgrounds, seasonal variants, and cleaner hero shots while keeping the product story consistent.

Try Product Edits

Marketing Asset Generation

Create launch banners, social visuals, and paid media concepts from one API route. GPT Image 2 fits teams that need fast prompt iteration, visible pricing, and a direct path from playground testing to production use.

Test Marketing Prompts

Reference-Based Image Editing

Send prompts with reference images when you need targeted edits instead of net-new generation. This is useful for workflows that need controlled changes, updated compositions, or repeated variations from the same source asset.

Test Image Editing

Why Use GPT Image 2 Through EvoLink

EvoLink turns GPT Image 2 into a cleaner buying decision: one route, visible pricing, playground validation, and an async workflow that fits production image jobs.

One API Route from Testing to Production

Start in the playground, validate prompt structure, and move into API integration without switching tools or rebuilding your request flow.

Pricing and Billing Visibility

This page brings together pricing, playground access, and the request model name so teams can evaluate cost before scaling image workloads.

Async Workflow for Real Jobs

GPT Image 2 on EvoLink uses an asynchronous task flow, which fits longer-running generation and editing work better than pretending every request behaves like an instant sync response.

How to Integrate GPT Image 2

Three steps to test the route, send requests, and retrieve async image results.

Create Your API Key

Sign in to EvoLink and generate one API key for playground testing and production requests. This is the same access path used across supported model routes.

Send a GPT Image 2 Request

Use the `gpt-image-2` model name, then pass your prompt, size, quality, and optional `image_urls` when you need reference-based editing or image-to-image workflows.

Poll the Task Result and Save the Asset

Image generation runs as an async task. Retrieve the task result, handle completion in your app workflow, and store the returned asset promptly because generated image links are time-limited.

Open Playground

GPT Image 2 Capabilities

The practical controls that matter when you are evaluating an image API for production use.

Workflow

Text-to-Image and Image Editing

Use one route for net-new image generation, reference-based edits, and image-to-image style workflows.

Editing

Reference Image Support

Attach one or more reference image URLs when your workflow needs controlled edits, variations, or prompt-guided changes from an existing asset.

Controls

Size and Quality Controls

Choose supported aspect ratios or pixel sizes, then adjust quality settings based on output requirements and cost tolerance.

Async

Async Task Delivery

Requests return task IDs for polling, which gives teams a cleaner way to manage longer-running generation and editing jobs.

Output

Single-Image Output Flow

The route is optimized around a single generated result per request, which simplifies downstream handling in approval, storage, and automation pipelines.

Delivery

Time-Limited Result URLs

Generated asset URLs are temporary, so production integrations should save completed outputs into app storage or a managed asset pipeline.

Cost Example

100 images (1K default)100 × $0.015 = $1.50

1,000 images (1K default)1,000 × $0.015 = $15

Production scale 10,000 images/month10,000 × $0.015 = $150

Use low quality for prompt iteration (0.11× cost), promote to high only for final output.

Explore more image generation models on EvoLink →

Frequently Asked Questions

Everything you need to know about the product and billing.

On EvoLink, GPT Image 2 is an image generation route for text-to-image, reference-based editing, and image-to-image workflows. This page is the product entry point for pricing, playground testing, and API access.

Pricing depends on the current route configuration shown on this page. Review the pricing section before launch so your team can validate cost against size, quality, and expected image volume.

Yes. You can submit prompts with optional `image_urls` when you need reference-based edits, controlled variations, or image-to-image style requests instead of net-new generation only.

Use `gpt-image-2` as the request model name for this route. Keep that model name consistent across playground testing, implementation, and production rollout.

Yes. GPT Image 2 runs through an async task flow on EvoLink. Your app should create the task, poll for completion, and then store or deliver the finished image once the result is ready.

Use the playground on this page to validate prompts, compare settings, and confirm whether the route fits your workflow before you wire it into production.

GPT Image 2 supports multiple aspect ratios and pixel sizes on EvoLink, along with quality controls you can tune for output needs and cost. Check the request parameters and pricing block on this page for the current supported options.

Use the API section on this page to review endpoint behavior, request shape, and async task handling. That is the best place to confirm implementation details before shipping.

Explore the GPT Image family

GPT Image 2 is the newest route in the GPT Image family on EvoLink. Use the family page when your next question is which route to choose, then compare GPT Image 1.5 or GPT Image 1 if you need a production-default or legacy baseline.

View GPT Image family Open GPT Image 1.5 Open GPT Image 1

API Reference

Select endpoint

Authentication

All APIs require Bearer Token authentication.

Header

Authorization: 
Bearer YOUR_API_KEY

Get API Key

POST

/v1/images/generations

Generate Image

Create an image generation task using text prompts. Supports text-to-image and reference-image-assisted generation.

Asynchronous processing mode, use the returned task ID to query status.

Generated image links are valid for 24 hours, please save them promptly.

Beta is billed per call, tiered by resolution. auto is billed at the 1K tier. When size is explicit WxH pixels, the tier is inferred from total pixels (≤ 1.7 MP → 1K, ≤ 4.4 MP → 2K, > 4.4 MP → 4K). n is fixed at 1; quality is not exposed on this channel.

Request Parameters

modelstringRequiredDefault: gpt-image-2-beta

Image generation model name.

Value	Description
gpt-image-2-beta	GPT Image 2 Beta — per-call billing tiered by resolution

Examplegpt-image-2-beta

promptstringRequired

Prompt describing the image to be generated or how to edit the reference image.

Notes

Max 32,000 characters (counted by Unicode code points — works for CJK and other languages)

ExampleA beautiful colorful sunset over the ocean

sizestringOptionalDefault: auto

Aspect ratio, `auto`, or explicit WxH pixels. With ratio, backend combines it with the resolution tier to compute the final pixels. With WxH, resolution is ignored and the billing tier is inferred from total pixels.

Value	Description
auto	Default — model decides; billed at the 1K tier
1:1	Square
1:2 / 2:1	Extreme portrait / landscape
1:3 / 3:1	Ultra portrait / landscape (3:1 limit)
2:3 / 3:2	Standard portrait / landscape
3:4 / 4:3	Classic portrait / landscape
4:5 / 5:4	Common social media
9:16 / 16:9	Mobile / desktop widescreen
9:21 / 21:9	Ultra-wide
1024x1024 (or any WxH)	Explicit pixels — multiples of 16, total 655K~8.29M, edges ≤ 3840, aspect ≤ 3:1

Example1:1

resolutionstringOptionalDefault: 1K

Resolution tier — drives both the final pixel size (combined with size ratio) and the per-call price. Ignored when size is `auto` or explicit WxH (auto is billed at 1K; WxH infers the tier from total pixels).

Value	Description
1K	Pixel budget ≈ 1024² = 1.05MP (1:1 → 1024×1024, 16:9 → 1360×768)
2K	Pixel budget ≈ 2048² = 4.19MP (1:1 → 2048×2048, 16:9 → 2736×1536)
4K	Pixel budget = 8.29MP / MaxPixels (1:1 → 2880×2880, 16:9 → 3840×2160 UHD)

Example1K

image_urlsarrayOptional

Reference image URL list for image-to-image and image editing features.

Notes

1–16 images per request
Each image ≤ 50 MB
Supported formats: .jpeg, .jpg, .png, .webp
URLs must be directly accessible by the server, or URLs that trigger direct download (typically URLs ending with image extensions like .png, .jpg)

Examplehttps://example.com/image1.png

callback_urlstringOptional

HTTPS callback address after task completion.

Notes

Triggered on completion, failure, or cancellation
Sent after billing confirmation
HTTPS only, no internal IPs
Max length: 2048 chars
Timeout: 10s, Max 3 retries

Examplehttps://your-domain.com/webhooks/image-task-completed

Request Example

{
  "model": "gpt-image-2-beta",
  "prompt": "A beautiful colorful sunset over the ocean",
  "size": "1:1",
  "resolution": "1K",
  "image_urls": [
    "https://example.com/reference.png"
  ]
}

Response Example

{
  "created": 1757156493,
  "id": "task-unified-1757156493-imcg5zqt",
  "model": "gpt-image-2-beta",
  "object": "image.generation.task",
  "progress": 0,
  "status": "pending",
  "task_info": {
    "can_cancel": true,
    "estimated_time": 180
  },
  "type": "image",
  "usage": {
    "billing_rule": "per_call",
    "credits_reserved": 1.02,
    "user_group": "default"
  }
}