Seedance 2.5 from ByteDance is officially available on Apiframe.

GPT Image 2 API Guide: Features, Pricing & Code

GPT Image 2 explained: reasoning mode, multi-image consistency, real pricing, and working code to start today.

Renaud Last updated November 3, 2026 July 5, 2026 · 7 min read Beginner
GPT Image 2 API Guide: Features, Pricing & Code

OpenAI released GPT Image 2 on April 21, 2026, as the model behind ChatGPT Images 2.0. It succeeds GPT Image 1.5 and completes OpenAI's move away from DALL-E 3. It is the first OpenAI image model that thinks before it draws, and that shows up in cleaner text, better consistency across a set of images, and more accurate instruction following. It is live on Apiframe now: one key, one endpoint, and the same billing and job limits you already use for every other model on the platform.

This guide covers what is actually new, one real limitation worth knowing before you build, and a walkthrough of generating with the model, including pricing you can plan a budget around.

What's New in GPT Image 2

It thinks before it draws. Before generating an image, the model can plan the layout, search the web for reference material, and check its own output. No earlier OpenAI image model had this step, and it is the single biggest change in the release.

Up to 8 images from one prompt. It produces a matching set of images in a single request, keeping characters and objects looking like themselves across all of them. Useful for storyboards and campaign sets that previously meant generating images separately and hoping they matched.

Text that comes out readable. Dense text and non-Latin scripts render cleanly, reliably enough for posters, packaging, menus, and interface labels. These are exactly the places where earlier models tended to produce letter-shaped nonsense.

Closer attention to the prompt. OpenAI describes it as a major step up in following detailed prompts and putting objects where you asked for them in a scene.

Editing that holds its context. You can keep editing the same image across several rounds of instructions, and faces, objects, and composition survive the changes rather than drifting with each pass.

One real limitation worth knowing upfront

OpenAI's own model supports a wide range of output sizes, but the gpt-image-2 endpoint on Apiframe does not expose a resolution or size parameter. You get three aspect ratios (1:1, 3:2, and 2:3), and resolution follows from the quality setting rather than being something you ask for. If your project needs a specific pixel size, or an aspect ratio outside those three, that control is not available here today. Worth knowing before you design around a fixed resolution target.

Getting Started on Apiframe

GPT Image 2 runs through the same API key and billing as every other model on Apiframe, including Flux 2 Pro, Nano Banana, and Kling. It is not a separate integration, just one more model on the unified API you may already have. Requests run in the background: submit a job, then either check back for the result or have it delivered to a webhook, which is an automatic callback to your server when the job finishes. No OpenAI account required.

If you are starting from scratch, sign up for free credits, grab an API key, and you can be generating within minutes. Our getting started guide covers account setup, the full parameter reference lives in the GPT Image 2 docs, and you can try the model in Apiframe Studio without writing any code. For a broader look at how image APIs work, see our AI image API guide for developers, and if you are still choosing a route to the model, we have compared the main GPT Image 2 API providers.

Full Code Walkthrough

1. Text-to-image

A prompt, a quality setting, and an aspect ratio are all you need to start. Here it is in curl:

bash
curl -X POST https://api.apiframe.ai/v2/images/generate \
  -H "X-API-Key: afk_your_api_key_here" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-2",
    "prompt": "Infographic explaining how photosynthesis works, bold sans-serif labels, light academic color palette",
    "gptImage2Params": {
      "quality": "high",
      "aspect_ratio": "3:2"
    }
  }'

The same call in Python:

python
import requests

response = requests.post(
    "https://api.apiframe.ai/v2/images/generate",
    headers={
        "X-API-Key": "afk_your_api_key_here",
        "Content-Type": "application/json",
    },
    json={
        "model": "gpt-image-2",
        "prompt": "Infographic explaining how photosynthesis works, bold sans-serif labels, light academic color palette",
        "gptImage2Params": {
            "quality": "high",
            "aspect_ratio": "3:2",
        },
    },
)
print(response.json())  # { "jobId": "...", "status": "QUEUED" }

2. Image editing and combination

Pass one or more input_images to edit an existing image, or to combine subjects and styles into a single output.

python
response = requests.post(
    "https://api.apiframe.ai/v2/images/generate",
    headers={"X-API-Key": "afk_your_api_key_here"},
    json={
        "model": "gpt-image-2",
        "prompt": "Change the red hat to a light-blue velvet beret. Preserve the subject's face, pose, and lighting.",
        "gptImage2Params": {
            "input_images": ["https://example.com/portrait.jpg"],
            "quality": "auto",
        },
    },
)

The model reads your input images at full detail automatically. There is no setting to turn that up or down, which is a change from GPT Image 1.5, where you controlled it yourself. More on that difference in GPT Image 1.5 vs 2: Which One to Use. If you are building an editing-focused workflow, our AI image editing API guide covers the wider landscape.

3. Poll for the result

The submission returns a jobId right away, before the image exists. Poll GET /v2/jobs/{id} until the status is COMPLETED, or supply a webhook_url instead.

python
import time

while True:
    job = requests.get(
        f"https://api.apiframe.ai/v2/jobs/{job_id}",
        headers={"X-API-Key": "afk_your_api_key_here"},
    ).json()
    if job["status"] in ("COMPLETED", "FAILED"):
        break
    time.sleep(3)

print(job["result"])  # { "images": ["https://cdn2.apiframe.ai/images/..."] }

Average completion time is around 15 seconds, quick enough that polling works fine for low volumes. Once you are running many jobs at once, webhooks scale better, and our walkthrough on setting up webhook callbacks covers the handler and retry logic.

Pricing

GPT Image 2 charges a flat credit cost per image, set by the quality setting, and the total is multiplied by number_of_images.

QualityCreditsPrice ($0.01/credit, every plan)
Low2$0.02
Medium7$0.07
High19$0.19
Auto19$0.19

Credits cost a flat $0.01 each on every plan, whether you top up as you go or subscribe monthly. Plans differ in how many credits you get each month and how many jobs you can run at once, not in the per-credit rate, so the per-image prices above hold whichever tier you are on. The pricing page has the full plan breakdown.

For example, a request with quality: "medium" and number_of_images: 4 costs 7 × 4 = 28 credits, or $0.28. To compare costs across image models more broadly, see AI image API pricing in 2026: what you actually pay per image.

GPT Image 2 vs GPT Image 1.5

GPT Image 2 is the default choice now, but gpt-image-1.5 is still available on Apiframe and still has one job. It is the version to use if you need images with transparent backgrounds, which GPT Image 2 dropped in favor of automatic or solid backgrounds only. Pricing is not a reason to choose it: the two cost the same at low and medium quality, and GPT Image 2 is actually one credit cheaper at high and auto. See the full breakdown of what changed in GPT Image 1.5 vs 2: Which One to Use.

Use Cases

Storyboards and brand campaigns. Because up to 8 images stay consistent with each other, multi-panel sequences and campaign sets become practical in one request instead of several that never quite match.

Posters, packaging, and interface mockups. Text comes out clean enough to ship rather than as a placeholder you redo by hand. Pair it with Flux 2 Pro if you want to compare photorealistic styles on the same project, and our Flux vs Stable Diffusion vs Midjourney comparison covers how those houses differ.

Product edits. Editing across several rounds keeps the subject and composition intact, so you can refine one asset instead of regenerating it from scratch every time.

Educational and editorial content. Infographics, diagrams, and blog covers benefit directly from the planning step, which works out the layout before rendering anything. For a worked example, see how to generate a blog cover with GPT Image 2.

If you are comparing image models more broadly before committing, see 8 Best AI Image Generator APIs in 2026 or Best AI Image APIs in 2026. If keeping a character recognizable across a series matters to your project, our guide on keeping AI-generated characters consistent goes deeper, and if you want better results from the same model, see how prompt wording shapes photorealistic output.

FAQ

What is GPT Image 2?

OpenAI's newest image generation and editing model, released April 21, 2026, as the engine behind ChatGPT Images 2.0. It succeeds GPT Image 1.5 and is the first in the line to plan an image before generating it.

How is it different from GPT Image 1.5?

It adds a planning step before generation, handles non-Latin text better, produces consistent sets of images, and drops transparent background support. Full comparison in GPT Image 1.5 vs 2: Which One to Use.

Does it support image editing?

Yes, through the input_images parameter. Pass one image to edit it, or pass several to combine their subjects or styles into a single output.

What resolutions and aspect ratios does it support on Apiframe?

Three aspect ratios (1:1, 3:2, and 2:3), with resolution following from the quality setting rather than being selectable. See the limitation noted above.

How much does the GPT Image 2 API cost?

2 credits ($0.02) per image at low quality, 7 ($0.07) at medium, and 19 ($0.19) at high or auto, multiplied by the number of images you request.

Do I need an OpenAI account?

No. Apiframe manages access on your behalf, using the same key and billing as every other model on the platform. Get your API key and start with free credits, or go straight to the GPT Image 2 model page for the full spec and live docs.

Power your next AI product with Apiframe.

Instant access to 70+ media models through a single API. Start free and scale when you're ready.