Image

GPT Image 2.5 API

GPT Image 2.5 API on APIMart gives you OpenAI's GPT Image 2.5 models, the API version of ChatGPT Images 2.5 released on September 8, 2026. Use gpt-image-2.5-flare, the default choice with 50% lower latency than GPT Image 2, or gpt-image-2.5-sunburst for tighter control across detailed edits. Both generate and edit images at up to 4K, accept up to 16 reference images and cost the same per token.

Flare & SunburstUp to 4KMask Editing16 References

Get API Key API Docs

Service Highlights

Official Discounts
Pay-as-you-go
Fast & Low Latency

Input

Model

Prompt

Resolution

Aspect Ratio

Output Format

Background

Moderation

Number of Images

Quality

Reference Image

Mask Image

Add at least one reference image when using a mask.

Output

Image link valid for 72 hours

GPT Image 2.5 API Pricing

Transparent pricing with no hidden fees. Pay only for what you use.

gpt-image-2.5-flare

Quality

Size

Resolution

Token Pricing

Text input40Credits / 1M tokens

Cached text input10Credits / 1M tokens

Image input64Credits / 1M tokens

Cached image input16Credits / 1M tokens

Image output240Credits / 1M tokens

Estimated Price

Save 20%

Showing: Gold member price

0.047Credits(~$0.0047)

0.0588Credits(~$0.0059)Original Price

Price is calculated from token usage — this configuration uses about 196 tokens.

* Estimated output cost for one image, excluding prompt and reference-image input costs. Final charges are based on actual token usage.

GPT Image 2.5 API - OpenAI Flare and Sunburst Image Models

Generate and edit images with GPT Image 2.5 on APIMart: pick Flare for fast everyday images or Sunburst for precise multi-turn edits, then output up to 4K with transparent backgrounds, mask inpainting and token billing at the same rate for both models.

50K+

Active Users

99.9%

Uptime

Faster

70%

Cost Savings

Key Features of the GPT Image 2.5 API

What gpt-image-2.5-flare and gpt-image-2.5-sunburst accept, return and cost on APIMart

GPT Image 2.5 Flare: 50% Lower Latency

OpenAI positions Flare as the default for most apps: higher-quality images than GPT Image 2 at 50% lower latency, suited to social content, product experiences, prototypes and high-volume generation.

GPT Image 2.5 Sunburst: Precision Editing

Sunburst gives tighter control across multi-turn edits for campaign creative and polished product imagery. It takes longer to generate than Flare.

Text to Image and Image Editing

Generate from a prompt, or add up to 16 reference images in image_urls. GPT Image 2.5 is better at changing only what you ask while keeping the subject and composition.

Mask Inpainting with mask_url

Pass a PNG mask with an alpha channel, the same size as the first input image. Fully transparent pixels mark the area to edit. Both Flare and Sunburst support masks.

Up to 4K in 15 Aspect Ratios

Choose 1k, 2k or 4k with ratios from 1:1 to 21:9 and 3:1, or send exact pixel sizes in multiples of 16 with a long side up to 3840 px.

Six Quality Levels, Including xhigh and max

Set quality to low, medium, high, xhigh, max or auto. xhigh and max are new in GPT Image 2.5, and medium and high use about a quarter of the output tokens of GPT Image 2.

Transparent Backgrounds and Three Formats

Output PNG, JPEG or WebP with optional compression. Set background to transparent with PNG or WebP for stickers, icons and cutouts.

One Token Price for Flare and Sunburst

The official channel bills actual token usage, and Flare and Sunburst use identical rates, so you choose between them on speed and precision, not price.

What You Can Build with the GPT Image 2.5 API

Workflows that use Flare speed, Sunburst precision and GPT Image 2.5 editing controls

Social and Creator Content at Volume

Use Flare to generate posts, thumbnails and covers quickly, with up to 4 images per request.

Product Photos That Keep the Product Intact

Add product photos as references and place them in new scenes while the label, shape and colors stay recognizable.

Campaign Creative with One Element Changed

Use Sunburst and a mask to swap a background, price or headline while the rest of the ad stays the same.

Transparent Stickers, Icons and Cutouts

Request a transparent background in PNG or WebP and drop the result straight into a layout.

Visual Search and Product Experiences

Generate on-demand images inside apps, such as try-on previews or room ideas, where Flare's lower latency matters.

Storyboards and Multi-Turn Concept Work

Refine the same character or scene over several rounds; GPT Image 2.5 keeps edits consistent across turns.

How to Get a GPT Image 2.5 API Key and Start Building

From sign-up to your first GPT Image 2.5 image in three steps

Create Your GPT Image 2.5 API Key

Register on APIMart, add balance and create an API key in the dashboard. The same key calls Flare, Sunburst and every other model.

→

Test Flare and Sunburst in the Playground

Try prompts, references, sizes and quality levels above and check the estimated cost before you write code.

→

Integrate the GPT Image 2.5 API

Submit tasks from your backend with the example below, then poll for the result.

GPT Image 2.5 API Example: Text to Image and Mask Editing

Call gpt-image-2.5-flare or gpt-image-2.5-sunburst with one POST request. Generation is asynchronous: submit a task, then poll the task endpoint for the image URL.

1
Authenticate GPT Image 2.5 API requests
Save your APIMart key in the APIMART_API_KEY environment variable and send it as Authorization: Bearer <key> on every request.
2
Submit a GPT Image 2.5 task
POST /v1/images/generations with model gpt-image-2.5-flare or gpt-image-2.5-sunburst, your prompt, size, resolution and quality. Add image_urls and mask_url to edit. The task ID is in data[0].task_id.
3
Get the GPT Image 2.5 result
Poll GET /v1/tasks/{task_id} every 2–5 seconds until status is completed, then read the image URL from data.result.images[0].url[0]. APIMart keeps results for 72 hours, so download the files you want to keep.
See the GPT Image 2.5 API reference for every parameter, size rule and response field.

POST /v1/images/generations

curl -X POST https://api.apimart.ai/v1/images/generations \
  -H "Authorization: Bearer $APIMART_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-2.5-flare",
    "prompt": "A minimalist poster for a jazz night in Lisbon, the words LISBON JAZZ in bold serif type, warm dusk light, film grain",
    "n": 1,
    "size": "16:9",
    "resolution": "2k",
    "quality": "high"
  }'

import os
import time
import requests

API_BASE = "https://api.apimart.ai"
headers = {"Authorization": f"Bearer {os.environ['APIMART_API_KEY']}"}

# 1. Submit the generation task
payload = {
  "model": "gpt-image-2.5-flare",
  "prompt": "A minimalist poster for a jazz night in Lisbon, the words LISBON JAZZ in bold serif type, warm dusk light, film grain",
  "n": 1,
  "size": "16:9",
  "resolution": "2k",
  "quality": "high"
}
resp = requests.post(f"{API_BASE}/v1/images/generations", headers=headers, json=payload, timeout=60)
resp.raise_for_status()
task_id = resp.json()["data"][0]["task_id"]

# 2. Poll the task until it finishes
for _ in range(200):
    task_resp = requests.get(f"{API_BASE}/v1/tasks/{task_id}", headers=headers, timeout=30)
    task_resp.raise_for_status()
    task = task_resp.json()["data"]
    if task["status"] == "completed":
        print(task["result"]["images"][0]["url"][0])
        break
    if task["status"] == "failed":
        raise RuntimeError(task.get("error", {}).get("message", "generation failed"))
    time.sleep(3)
else:
    raise TimeoutError(f"task {task_id} did not finish in time")

curl -X POST https://api.apimart.ai/v1/images/generations \
  -H "Authorization: Bearer $APIMART_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-2.5-sunburst",
    "prompt": "Replace only the masked background with a soft pastel studio backdrop and keep the product and its label unchanged",
    "image_urls": [
      "https://example.com/product.png"
    ],
    "mask_url": "https://example.com/product-mask.png",
    "resolution": "2k",
    "quality": "high"
  }'

TASK_ID="task_xxx"  # replace with the task_id returned by the POST request
curl "https://api.apimart.ai/v1/tasks/$TASK_ID" \
  -H "Authorization: Bearer $APIMART_API_KEY"

GPT Image 2.5 API Parameters

Parameter	Type	Required	Default	Description
`model`	string	Required	Default: gpt-image-2.5-flare	gpt-image-2.5-flare or gpt-image-2.5-sunburst. Both take the same parameters.
`prompt`	string	Required	Default: —	What to generate or change: subject, scene, composition, style, lighting and anything that must stay the same.
`size`	string	Optional	Default: auto	auto, one of 15 aspect ratios (1:1, 3:2, 2:3, 4:3, 3:4, 5:4, 4:5, 16:9, 9:16, 2:1, 1:2, 21:9, 9:21, 3:1, 1:3) or exact pixels such as 1600x1200. Omit it when editing to follow the input ratio.
`resolution`	string	Optional	Default: 1k	1k, 2k or 4k, used with an aspect-ratio size. Ignored for exact pixel sizes.
`quality`	string	Optional	Default: auto	low, medium, high, xhigh, max or auto. auto reserves funds at max and settles on actual usage.
`n`	integer	Optional	Default: 1	Number of images, 1 to 4.
`output_format`	string	Optional	Default: png	png, jpeg or webp.
`output_compression`	integer	Optional	Default: —	0 to 100, for jpeg and webp only.
`background`	string	Optional	Default: —	transparent, opaque or auto. transparent needs png or webp.
`moderation`	string	Optional	Default: low	auto or low.
`image_urls`	array	Optional	Default: —	Up to 16 public HTTP(S) reference image URLs. Adding them switches to editing.
`mask_url`	string	Optional	Default: —	PNG mask with an alpha channel (URL or PNG data URL), the same size as the first input image. Transparent pixels mark the area to edit.

GPT Image API Model Family

Compare every GPT Image model on APIMart. One API key calls all of them; GPT Image 2.5 Flare and Sunburst share the same parameters and pricing.

Model	Best for	Resolution / size
GPT Image 2.5 APIThis page	Flare and Sunburst on one page, with up to 16 reference images, masks and transparent backgrounds	1K / 2K / 4K
GPT Image 2.5 Flare API	Fast everyday images, batch generation and rapid prototyping	1K / 2K / 4K
GPT Image 2.5 Sunburst API	Precise edits for production assets, ad creatives and multi-turn editing	1K / 2K / 4K
GPT Image 2 API	Text-to-image and editing with up to 16 reference images and 15 aspect ratios	1K / 2K / 4K
GPT Image 1.5 API	Text-to-image and editing with up to 5 reference images and masks	1:1 / 3:2 / 2:3
GPT Image 1 API	Previous generation: text-to-image and editing with up to 5 reference images and masks	1:1 / 3:2 / 2:3

GPT Image 2.5 API FAQ

Answers about GPT Image 2.5, Flare vs Sunburst, parameters, pricing and the two APIMart channels

What is GPT Image 2.5?

GPT Image 2.5 is OpenAI's image model family for generation and editing, released on September 8, 2026 alongside ChatGPT Images 2.5. The API offers two models, gpt-image-2.5-flare and gpt-image-2.5-sunburst, and APIMart serves both under those IDs. OpenAI writes the name as GPT-Image-2.5.

When was GPT Image 2.5 released?

OpenAI announced ChatGPT Images 2.5 and the GPT Image 2.5 Flare and Sunburst API models on September 8, 2026. Both models are available through the APIMart GPT Image 2.5 API now.

Is GPT Image 2.5 the same as ChatGPT Images 2.5?

They share the same model improvements. ChatGPT Images 2.5 is the image feature inside ChatGPT, with tools such as Sketch and templates, while GPT Image 2.5 Flare and Sunburst are the API models you call from your own code.

GPT Image 2.5 Flare vs Sunburst: which should I use?

Start with Flare, OpenAI's default for most apps, when speed and volume matter. Choose Sunburst for production ads, product imagery and detailed multi-turn edits that need tighter control, and expect longer generation times. Both cost the same and accept the same parameters.

How is GPT Image 2.5 different from GPT Image 2?

Flare delivers higher-quality images than GPT Image 2 at 50% lower latency, and GPT Image 2.5 keeps reference subjects and edits more consistent. It adds the xhigh and max quality levels, and its medium and high levels use about a quarter of the output tokens of GPT Image 2.

What parameters does the GPT Image 2.5 API support?

model and prompt are required. Optional fields are size (auto, 15 aspect ratios or exact pixels), resolution (1k, 2k or 4k, default 1k), quality (low, medium, high, xhigh, max or auto), n (1–4), output_format (png, jpeg or webp), output_compression, background, moderation, image_urls (up to 16) and mask_url.

Can GPT Image 2.5 edit images with references and masks?

Yes. Add up to 16 public image URLs in image_urls to edit or combine images. For local inpainting, add mask_url: a PNG with an alpha channel, the same size as the first input image, where fully transparent pixels mark the area to change. Keep masks under 4 MB.

What sizes and resolutions does GPT Image 2.5 support?

Use auto or one of 15 aspect ratios at 1k, 2k or 4k, for example 1:1 at 2880×2880 or 16:9 at 3840×2160 in 4k. Exact sizes must be multiples of 16, at most 3840 px per side, no wider than 3:1 and between 655,360 and 8,294,400 pixels. Sizes above 2560×1440 are experimental.

Does GPT Image 2.5 support transparent backgrounds?

Yes. Set background to transparent with output_format png or webp. JPEG has no alpha channel, so transparent JPEG requests are not supported.

How do I call the GPT Image 2.5 API and get the image?

POST /v1/images/generations and read the task ID from data[0].task_id. Poll GET /v1/tasks/{task_id} every 2–5 seconds until status is completed or failed, then read data.result.images[].url[]. Results are kept for 72 hours, and failed tasks are refunded.

How much does the GPT Image 2.5 API cost?

On the official channel GPT Image 2.5 is billed by actual token usage for text input, image input and image output, and Flare and Sunburst use identical rates. Cost rises with quality and size; with quality auto, funds are reserved at max and settled on actual use. Gold members save 20%, and Platinum and Diamond members save up to 28%. See the pricing table on this page.

What is the difference between the official and Ext channels?

The official channel uses gpt-image-2.5-flare or gpt-image-2.5-sunburst with token billing, 15 aspect ratios, exact pixel sizes, quality levels, formats and masks. The Ext channel uses gpt-image-2.5-ext with version flare or sunburst, 10 aspect ratios or auto at 1K, 2K or 4K, and a price per delivered image by version and resolution, with reference images free.

Is GPT Image 2.5 free or open source?

No. APIMart has no free quota for GPT Image 2.5; you pay as you go for each task. OpenAI offers GPT Image 2.5 only as a hosted API and has not released its weights, so it cannot be downloaded or run locally.

How do I get a GPT Image 2.5 API key?

Sign up on APIMart, add balance and create a key on the API Keys page. Send it as Authorization: Bearer <key>. The same key calls GPT Image 2.5 Flare and Sunburst, GPT Image 2 and every other model on APIMart.

Related Models

Explore more models in the same category.

View All Models →

GPT Image 2.5 API FAQ

Answers about GPT Image 2.5, Flare vs Sunburst, parameters, pricing and the two APIMart channels

What is GPT Image 2.5?

When was GPT Image 2.5 released?

OpenAI announced ChatGPT Images 2.5 and the GPT Image 2.5 Flare and Sunburst API models on September 8, 2026. Both models are available through the APIMart GPT Image 2.5 API now.

Is GPT Image 2.5 the same as ChatGPT Images 2.5?

GPT Image 2.5 Flare vs Sunburst: which should I use?

How is GPT Image 2.5 different from GPT Image 2?

What parameters does the GPT Image 2.5 API support?

Can GPT Image 2.5 edit images with references and masks?

What sizes and resolutions does GPT Image 2.5 support?

Does GPT Image 2.5 support transparent backgrounds?

Yes. Set background to transparent with output_format png or webp. JPEG has no alpha channel, so transparent JPEG requests are not supported.

GPT Image 2.5 API

GPT Image 2.5 API Pricing

GPT Image 2.5 API - OpenAI Flare and Sunburst Image Models

Key Features of the GPT Image 2.5 API

What You Can Build with the GPT Image 2.5 API

How to Get a GPT Image 2.5 API Key and Start Building

Create Your GPT Image 2.5 API Key

Test Flare and Sunburst in the Playground

Integrate the GPT Image 2.5 API

GPT Image 2.5 API Example: Text to Image and Mask Editing

Authenticate GPT Image 2.5 API requests

Submit a GPT Image 2.5 task

Get the GPT Image 2.5 result

GPT Image 2.5 API Parameters

GPT Image API Model Family

GPT Image 2.5 API FAQ

What is GPT Image 2.5?

When was GPT Image 2.5 released?

Is GPT Image 2.5 the same as ChatGPT Images 2.5?

GPT Image 2.5 Flare vs Sunburst: which should I use?

How is GPT Image 2.5 different from GPT Image 2?

What parameters does the GPT Image 2.5 API support?

Can GPT Image 2.5 edit images with references and masks?

What sizes and resolutions does GPT Image 2.5 support?

Does GPT Image 2.5 support transparent backgrounds?

How do I call the GPT Image 2.5 API and get the image?

How much does the GPT Image 2.5 API cost?

What is the difference between the official and Ext channels?

Is GPT Image 2.5 free or open source?

How do I get a GPT Image 2.5 API key?

Related Models

Key Features of the GPT Image 2.5 API

What You Can Build with the GPT Image 2.5 API

How to Get a GPT Image 2.5 API Key and Start Building

Create Your GPT Image 2.5 API Key

Test Flare and Sunburst in the Playground

Integrate the GPT Image 2.5 API

GPT Image 2.5 API FAQ

What is GPT Image 2.5?

When was GPT Image 2.5 released?

Is GPT Image 2.5 the same as ChatGPT Images 2.5?

GPT Image 2.5 Flare vs Sunburst: which should I use?

How is GPT Image 2.5 different from GPT Image 2?

What parameters does the GPT Image 2.5 API support?

Can GPT Image 2.5 edit images with references and masks?

What sizes and resolutions does GPT Image 2.5 support?

Does GPT Image 2.5 support transparent backgrounds?

How do I call the GPT Image 2.5 API and get the image?

How much does the GPT Image 2.5 API cost?

What is the difference between the official and Ext channels?

Is GPT Image 2.5 free or open source?

How do I get a GPT Image 2.5 API key?

Related Models