GPT Image 2 — Image Generation & Editing
GPT Image 2 generates images from text and edits reference images with optional masks. Create transparent backgrounds, choose your output size and quality, and try it directly in the Playground.
- Text-to-image
- Batch up to 10
- Task API
- CDN URLs
- Usage-based billing
- No subscription
Try GPT Image 2 before you write any code
Generate or edit images with the same parameters and asynchronous API used by your integration.
gpt-image-2-officialFrom $22.50 / 1M output tokens
Generate from a prompt or edit reference images. Submit once; results appear when the task completes.
required · up to 32,000 characters
Reference images for image-to-image and editing. Files are uploaded and sent as URLs; each one adds image input tokens.
Let the model choose the image dimensions.
auto or WIDTHxHEIGHT · custom sizes supported
Reference images for image-to-image and editing. Files are uploaded and sent as URLs; each one adds image input tokens.
PNG smaller than 4 MB, matching the first reference image dimensions. Transparent areas are edited.
default auto · biggest cost lever
1 to 10 · default 1
auto, opaque or transparent · transparency requires PNG
default png
Choose JPEG to set compression.
default auto
Greyed-out options relax content filtering. They are available through the API only, not in the playground.
Reference estimate using a 1024×1024 sample at the selected quality and image count, priced at current rates. Auto uses a measured auto sample. Actual size, prompt and reference images can change the final cost.
Try one of these prompts
Generated with gpt-image-2 through this API.
What GPT Image 2 costs
Every figure below is computed from the rate that actually bills you, read live when the page loads. If that rate ever changes, these numbers change with it.
- Official
- $0.00448/image
- Standard
- $0.0315/image
- Official
- $0.04417/request
- Standard
- $0.315/request
- Official
- 4,466images
- Standard
- 634images
On Official IDs, size and quality change token usage, so cost is not a fixed function of pixel dimensions. Standard IDs charge one flat price per delivered image. GPT Image 2 charges nothing when a request fails or comes back without an image.
Measured on real renders through the Official channel: 196 image output tokens for one 1024×1024 image at low quality, plus 18 tokens for the prompt itself. A longer prompt spends more input tokens.
Measured cost per quality tier
One 1024×1024 image, same prompt, one render per tier. Output tokens are what the model reported on the Official channel, priced at its current rate; Standard charges the same flat price at every tier.
| Quality | Output tokens | Cost per image · Official | Cost per image · Standard |
|---|---|---|---|
| low | 196 | $0.00448 | $0.0315 |
| medium | 1,756 | $0.03958 | $0.0315 |
| high | 7,024 | $0.15811 | $0.0315 |
| auto (model picks) | 186 | $0.00425 | $0.0315 |
Measured once per tier, not averaged. A render of a different subject at the same settings will land near these numbers, not exactly on them.
Price your own request
Move the controls and watch the figure. It is computed from the live rate and the token counts measured below, so it changes when the price does.
A pre-submit estimate. Standard bills the images delivered; Official bills the text-input and image-output usage the model reports.
| Item | Rule | Rate | Billed as |
|---|---|---|---|
| Official · Image output tokens | Output token usage depends on image size and quality. | $22.50 / 1M$30.00 / 1M | Output tokens |
| Official · Text input tokens | The tokens your prompt spends. | $3.75 / 1M$5.00 / 1M | Input tokens |
| Standard · Per image | One flat price per delivered image, at any size or quality. | $0.0315 /image$0.042 /image | Images delivered |
| Failed requests | A request that returns no image. | Free | Not billed |
GPT Image 2 overview
Generate product shots, UI concepts and campaign variants with GPT Image 2 through SeedRouter. Start from a text prompt, pick a size and a quality tier, and get back image URLs you can download.
GPT Image 2 is OpenAI's image model for everyday generation. Calling it here is a task: you POST a prompt, receive a task id, and poll that id until the images are ready. Nothing holds your HTTP connection open while a render takes its time.
Best fit: text-to-image for product, marketing and UI work · Cost basis: prompt length, image size and quality tier · Output boundary: up to 10 images per request, delivered as CDN URLs
Use GPT Image 2 your way
Call it from your own server for product features and batch jobs, or drive it from a coding agent for one-off creative work. Both use the same key, the same balance and the same task history.
Integrate it in your backend
Call the image endpoint from your server and keep control of the prompt, the parameters, the queue and where the results are stored.
- 1Try a real brief in the playground above and check the cost
- 2Create an API key in your dashboard
- 3POST the prompt and keep the task id you get back
- 4Poll the task id and download the image URLs it returns
Drive it from a coding agent
Give an agent the key and the endpoint and let it write the request, wait for the task and save the files, instead of hand-coding each step.
- 1Put the key in your environment as SEEDROUTER_API_KEY, never in a file you commit
- 2Paste the prompt below into Claude Code, Codex or Gemini CLI
- 3Fill in the subject, size, quality and how many images you want
- 4Approve the estimated cost, then let the agent submit, poll and download
What GPT Image 2 gives you
GPT Image 2 turns a text prompt into finished images and hands back URLs you can download. You send an OpenAI-shaped request body, get a task id back immediately, and poll it until the images are ready. Nothing stays open on your connection while a render takes its time.
A request body you already know
Model, prompt, n, size, quality. If you have called an image endpoint before, you have already written your first GPT Image 2 request.
Submit once, get results asynchronously
GPT Image 2 returns a task id right away. Poll it when you like. No long-held HTTP connections to babysit.
Downloadable URLs, not base64 blobs
Finished images land on a CDN. GPT Image 2 gives you links you can hand straight to a browser or a job queue.
Up to 10 images in one call
Ask GPT Image 2 for a batch when you want options. You pay for the images that come back, not for the number you asked for.
Usage you can check
Every finished task reports its input and output tokens, so you can audit any GPT Image 2 charge against the response that produced it.
Credits that never expire
Top up once and spend it whenever. No subscription, no monthly minimum, no plan to outgrow.
What teams build with GPT Image 2
Product and marketing visuals
Generate hero shots and campaign variants from a brief. One GPT Image 2 call per variant, same pipeline.
UI and concept mockups
Fill a design draft with plausible imagery before a photo shoot exists. One GPT Image 2 call per placeholder.
Catalog and listing art
Drive GPT Image 2 from your catalog and get consistent imagery across a long list of items, without booking a studio.
Social variants at volume
One idea, ten framings. Ask GPT Image 2 for a batch and post the best.
Editorial and blog art
One GPT Image 2 call illustrates a post in the time it takes to write the headline.
Image features inside your app
Let your own users generate. The task id makes it easy to show progress in your UI.
Calling GPT Image 2 takes four steps
- 01
Create a key
Sign in, open the dashboard, create a key. The same key calls GPT Image 2 and every other model you have access to.
- 02
Add credits
Top up once. Credits do not expire and there is no plan attached to them.
- 03
Submit a prompt
POST your prompt to GPT Image 2. You get a task id back immediately.
- 04
Poll and download
GET the task until GPT Image 2 reports completed, then download the image URLs it returns.
Choose an image model for the job
Everything on this table is read from the same catalogue that bills you. Models SeedRouter does not serve are not listed.
| Feature | GPT Image 2 |
|---|---|
| Billing model | Per token of actual usage |
| Input | Text and reference images |
| Output | Up to 10 image URLs per request |
| Request shape | OpenAI-style: model, prompt, n, size, quality |
| Failed request | Not charged |
Other image models will appear here as they are added to the catalogue.
GPT Image 2 limits worth planning for
Text inside an image still needs review
Words rendered into an image can come back misspelled or mangled. Check anything with type on it before it ships.
Editing uses reference image URLs
Provide up to 16 reference images and an optional PNG mask. The mask must match the first image dimensions; transparent pixels identify the area to edit.
Result URLs are a handoff, not storage
Download what you intend to keep and put it wherever your product already stores assets.
A batch takes longer than a single image
Asking for ten images is one request but ten renders. Poll on a schedule instead of blocking a user on it.
Why call GPT Image 2 through SeedRouter
One key for every model
The key that calls GPT Image 2 calls the rest of the catalog too. One integration, one balance.
Billing you can check
Every GPT Image 2 charge traces back to the usage in the response, and the price on the site is the price that bills.
No subscription
Buy credits, spend them whenever. Nothing expires and nothing renews.
Failures are free
A GPT Image 2 request that produces no image produces no charge.
The bill follows the response
GPT Image 2 charges are computed from the token usage the model reports, not from the size of your request.
Nothing rendered, nothing charged
If a request ends without an image, it costs nothing. That is the one reliability promise worth making.
Short batches are billed short
Ask GPT Image 2 for ten and get seven, and you pay for seven.
Size and quality drive the price
Both change how many output tokens a GPT Image 2 render spends. Prototype at low quality, ship at the quality you need.
Images live on a CDN
Download what you want to keep. GPT Image 2 hands you a URL, not permanent storage.
Other image models on SeedRouter
Same key, same balance, same task API. Read live from the catalogue, so nothing here is a model you cannot actually call.
Generate and edit images with GPT Image 2.5 Flare or Sunburst on SeedRouter. Six quality steps up to max, transparent backgrounds, a live playground and pay-as-you-go pricing.
View pricingCreate and edit images with Nano Banana 2 (Gemini 3.1 Flash Image): 512px to 4K, 14 aspect ratios, up to 14 references. Pay per image or per token.
View pricingCreate and edit images with Nano Banana Pro (Gemini 3 Pro Image): 1K to 4K output, built-in thinking, up to 14 references. Pay per image or per token.
View pricingCreate and edit images with Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image), Google's fastest and cheapest image model: 1K output, 14 aspect ratios.
View pricingGPT Image 2 guides
Practical guides to costs, prompts, batch requests, and integration.
Get a GPT Image 2 API key, store it safely, add credits, and make a first request with curl, Python or JavaScript, plus the errors a new key most often hits.
The GPT Image 2 API parameters that shape the image: model name, size and resolution, aspect ratios, 4K limits, quality, formats and the errors they return.
A complete GPT Image 2 API example in Python: submit a request, poll the task, download the images to disk, edit with references and handle errors safely.
Let Codex, Claude Code or another coding agent make images with GPT Image 2 through your API key, with a reusable prompt that asks before every paid request.
Submit a set of GPT Image 2 prompts, save task IDs, resume polling after interruptions, and download completed images without resubmitting them.
Estimate GPT Image 2 costs from current SeedRouter rates, understand sample token usage, and budget for drafts, revisions, and final images.
Write clearer GPT Image 2 prompts, specify exact text, preserve details in reference edits, and revise one visual constraint at a time.
Migrate a GPT Image 2 integration to SeedRouter by mapping request fields, handling asynchronous tasks, and validating URL-based image delivery.
Common questions
What is GPT Image 2?+
GPT Image 2 is an image model you call over HTTP: you POST a prompt, get a task id, then poll that id until the images are ready and download the URLs it returns.
How much does GPT Image 2 cost?+
It depends on the channel. Official IDs bill the tokens each render reports: a 1024×1024 image at low quality measures about $0.00448. Standard IDs charge a flat $0.0315 per delivered image at any size or quality. Every figure on this page is computed from the rate that bills you.
Do failed GPT Image 2 requests cost anything?+
No. If a request ends without an image, nothing is charged. Ask for ten images and receive seven, and you are billed for seven.
Why does GPT Image 2 return a task id instead of the image?+
Because a render takes longer than an HTTP request should stay open. The task id lets you poll on your own schedule and show progress in your own UI.
How many images can one GPT Image 2 request return?+
Up to ten. You are billed for the images that come back, which makes batching a cheap way to get options.
Do credits expire?+
No. GPT Image 2 credits are a one-time purchase with no subscription attached. Top up when you need to and spend the balance whenever you like.
Will my existing image code work with GPT Image 2?+
Use the official GPT Image 2 parameters with asynchronous submission and polling. Submit once, save the task id, then poll for image URLs and token usage. An official synchronous or streaming SDK call needs this polling step added.
How long do GPT Image 2 image URLs stay up?+
Treat them as a handoff. Download anything you intend to keep and store it wherever your product already stores assets.
Start building with GPT Image 2
Try your first prompt in the Playground, or explore the API docs to add image generation to your app.