Cookie Preferences
We use cookies to personalize content and analyze traffic. See our Cookie Policy.
gpt-image-2.5
Save 94.9%
Compared with OpenAI Flare xhigh scenario (3840x2160, output only), for the same unit shown below.
Your Enterprise price
$0.01
USD · per image
GPT Image 2.5 4K
Unlock with a $5,000 top-up
Maintain Enterprise pricing with at least $5,000 in successful payments in every rolling 30-day period.
Top up & unlock EnterpriseNo image yet. Submit the form to generate image.
Compare this model across platforms in one place. ApiPass Enterprise pricing is highlighted for each configuration.
| Platform | Price | Enterprise savings |
|---|---|---|
| ApiPass Enterprise | $0.01 | Your Enterprise price |
| OpenAI Flare xhigh scenario (3840x2160, output only) | $0.18 | Save 94.9% |
| Platform | Price | Enterprise savings |
|---|---|---|
| ApiPass Enterprise | $0.01 | Your Enterprise price |
| OpenAI Flare xhigh scenario (2560x1440, output only) | $0.10 | Save 92.6% |
| Platform | Price | Enterprise savings |
|---|---|---|
| ApiPass Enterprise | $0.01 | Your Enterprise price |
| OpenAI Flare xhigh scenario (1280x720, output only) | $0.05 | Save 88.3% |
Each comparison uses the currency, billing unit and configuration shown. Savings compare ApiPass Enterprise with each platform; provider prices may change.
Explore different use cases and parameter configurations
Transparent pricing with no hidden fees. Pay as you go.
| Rule & Modality | Channel | Credits | Price (USD) | Official / Reference Price | Daily Savings |
|---|---|---|---|---|---|
GPT Image 2.5 4K imageOpenAIresolution: 4k / 4K | Starter | 0.2per image | $0.001 | Based on 10,000 images/dayOpenAI Flare xhigh scenario (3840x2160, output only) costs $1769.910/day moreApiPass is 99.49% lower | |
GPT Image 2.5 2K imageOpenAIresolution: 2k / 2K | Starter | 0.16per image | $0.001 | Based on 10,000 images/dayOpenAI Flare xhigh scenario (2560x1440, output only) costs $975.530/day moreApiPass is 99.26% lower | |
GPT Image 2.5 1K imageOpenAIresolution: 1k / 1K | Starter | 0.13per image | $0.001 | Based on 10,000 images/dayOpenAI Flare xhigh scenario (1280x720, output only) costs $498.990/day moreApiPass is 98.83% lower |
Competitor pricing costs $498.990/day more
ApiPass is 98.83% lower, estimated at Based on 10,000 images/day
GPT Image 2.5 1K
resolution: 1k / 1K
ApiPass Price
$0.001
0.13 credits per image
Competitor Prices
Complete guide to using Gpt Image 2.5
Build image generation and editing workflows with GPT Image 2.5 on ApiPass. Turn text prompts and reference images into product photos, campaign visuals, portraits, design assets, and precise creative revisions while preserving recognizable subjects and visual details.
GPT Image 2.5 is OpenAI’s image generation and editing model family, with Flare and Sunburst variants. Both accept text and image inputs and produce image outputs. The model family supports text-to-image generation, reference-led image-to-image editing, inpainting, high-fidelity subject consistency, complex layout following, transparent-background asset creation, and multi-turn creative revisions. Flare focuses on fast everyday generation and high-volume iteration, while Sunburst is designed for precision-focused image editing and detailed production assets.
Create product shots, campaign concepts, portraits, social graphics, illustrations, and presentation visuals from structured natural-language prompts. Describe subject placement, style, lighting, composition, and required text to guide the final image.
Use reference images together with text instructions to create new scenes while preserving recognizable people, products, facial traits, lighting, textures, and brand details. This workflow is ideal for portrait variations, e-commerce visuals, and branded creative assets.
Provide an existing image and describe the product, background element, copy, or object to update. GPT Image 2.5 supports targeted edits while preserving the surrounding composition and visual treatment.
Works from reference photos to re-create people, places, and objects across new scenes and styles while keeping them recognizable. Lighting and textures feel natural, and distinctive traits are preserved—making reference-driven workflows more dependable and keeping variations tied to the source.
The GPT Image 2.5 API makes pinpoint edits to exactly what you ask for and leaves everything else unchanged, enabling single-element swaps while preserving composition and brand treatment.
In multi-turn editing, the GPT Image 2.5 API keeps earlier adjustments consistent and stacks new ones without quality loss, supporting reliable, incremental production updates.
Faced with complex briefs, the GPT Image 2.5 API parses detailed visual instructions, honors requested styles, handles real-world context and complex/transparent layouts, and keeps outputs aligned with your direction.
To speed creative starts, the GPT Image 2.5 API offers prompt templates (e.g., Poster, Merch) so you can drop in content, design cues, and style choices instead of starting from a blank canvas.
When rough drawings convey ideas faster than words, the GPT Image 2.5 API ingests sketches to define layout, contours, placement, and proportions, then refines results as you add guidance.
To speed creative starts, the GPT Image 2.5 API offers prompt templates (e.g., Poster, Merch) so you can drop in content, design cues, and style choices instead of starting from a blank canvas.
Use product reference photos to create new backgrounds, seasonal compositions, listing images, and polished product scenes while retaining the product’s defining details.
Develop campaign concepts with fast generation, then refine selected visuals with precise edits. Create variations in composition, background, and visual treatment around a consistent creative brief.
Generate reference-led portraits, promotional graphics, themed posts, and visual variations. Explore ideas quickly and make focused changes to a selected image without restarting the composition.
Translate layout instructions, artistic direction, and reference assets into illustrations, presentation graphics, educational visuals, and publication-ready image concepts.
Build application workflows around text prompts, reference inputs, asynchronous task submission, status polling, webhook callbacks, generation, and localized editing.
GPT Image 2 and GPT Image 2.5 both generate and edit images from text and image inputs. GPT Image 2.5 builds on that workflow with stronger reference fidelity, more precise editing, better multi-turn consistency, and faster generation through Flare.
GPT Image 2.5 improves preservation of recognizable subjects, lighting, facial traits, textures, and product details compared with GPT Image 2’s reference-led generation and editing workflow.
GPT Image 2.5 provides stronger targeted editing with improved preservation of surrounding details, making it better suited for product retouching, background revisions, and controlled inpainting tasks.
GPT Image 2.5 improves retention of earlier edits and image quality across multiple turns. Use each result as the next reference to refine creative direction in stages.
OpenAI reports that GPT Image 2.5 Flare reduces generation latency by 50% compared with GPT Image 2. Sunburst remains focused on precision and may take longer for detailed production editing.
Submit image generation or editing jobs through the create-task endpoint, receive a task ID, and query status until the task reaches success or fail. Completed results are returned in resultJson with generated image URLs.
Provide callBackUrl to receive task completion notifications automatically. If callBackUrl is omitted, no callback will be sent and your application can poll the query endpoint instead.
Credit consumption is based on output resolution in the current API workflow: 1K costs 10 credits, 2K costs 15 credits, and 4K costs 22 credits. Higher resolutions can provide more detail but may take longer to generate.
Use the channel parameter when you need a specific ApiPass provider route, or keep the default auto value to let ApiPass allocate tasks based on real-time pricing and stability metrics.
Describe the subject, composition, lighting, style, and required text. For edits, provide the source image and name the elements to change or preserve so the model can keep identity, layout, and visual treatment consistent.
Choose the aspect ratio, resolution, and output format supported by the API. Use reference images when you need image-to-image generation, precise editing, subject consistency, or multi-turn revision workflows.
Send a POST request to /api/v1/jobs/createTask with your input object. Store the returned taskId, then poll /api/v1/jobs/recordInfo or use callBackUrl to receive the result when generation completes.
All APIs require authentication via Bearer Token.
Authorization: Bearer
Submit a new GPT Image 2.5 image generation or image editing task.
The API accepts a JSON payload with the following structure:
1{
2 "model": "openai/gpt-image-2.5",
3 "callBackUrl": "string (optional)",
4 "channel": "auto",
5 "input": {
6 "prompt": "string",
7 "input_urls": ["string"],
8 "aspect_ratio": "auto",
9 "resolution": "1k"
10 }
11}modelRequiredstringThe model name to use for generation or editing. openai/gpt-image-2.5 uses GPT Image 2.5; on the KIE channel it routes to the Flare upstream model.
Available options:
"openai/gpt-image-2.5"
callBackUrlOptionalstringCallback URL for task completion notifications. If omitted, no callback will be sent.
"https://your-domain.com/api/callback"
channelOptionalstringYou may specify the corresponding provider within APIPASS via the channel parameter; these providers handle the actual image and video generation tasks. APIPASS currently offers three provider options:
The default value for the channel parameter is auto. When enabled, APIPASS automatically allocates tasks across available providers based on real-time pricing and stability metrics to balance minimal cost and reliable performance. Retain the default auto value unless you have custom routing requirements.
Available options:
auto
All GPT Image 2.5 models use the same input parameters. The input object contains the following fields:
input.promptRequiredstringText prompt describing what to generate or modify. Required for both text-to-image and image-to-image modes. Must be a non-empty string from 1 to 20,000 characters.
Describe the subject, composition, lighting, style, and any required edits. For image-to-image tasks, explain how the reference image should be modified and what should be preserved.
"A cinematic portrait of a red fox in a snowy forest."
input.input_urlsOptionalarray[string]List of reference image URLs. Supports up to 16 images. If the array is empty or omitted, the task runs in text-to-image mode; if the array contains one or more URLs, the task runs in image-to-image mode.
Provide at least one image URL to switch automatically to image-to-image mode. Use publicly accessible https:// URLs whenever possible. prompt is still required when reference images are provided.
["https://example.com/source.png"]
input.aspect_ratioOptionalstringOutput aspect ratio. Defaults to auto. Supported values are auto, 1:1, 3:2, 2:3, 16:9, 9:16, 4:3, 3:4, 21:9, 27:16, 16:27, 9:8, and 8:9.
If an upstream provider does not support the exact requested aspect ratio, the adapter maps it to the closest supported upstream size tier.
Available options:
"16:9"
input.resolutionOptionalstringOutput resolution tier. Defaults to 1k. Supported values are 1k, 2k, and 4k.
Values are case-insensitive. resolution affects upstream model selection and may be converted to provider-specific fields or model suffixes.
Available options:
"1k"
1curl -X POST "https://api.apipass.dev/api/v1/jobs/createTask" \
2 -H "Authorization: Bearer YOUR_APIPASS_API_KEY" \
3 -H "Content-Type: application/json" \
4 -d '{
5 "model": "openai/gpt-image-2.5",
6 "callBackUrl": "https://your-domain.com/api/callback",
7 "channel": "auto",
8 "input": {
9 "prompt": "A premium product photo of a silver espresso machine in a bright studio.",
10 "aspect_ratio": "1:1",
11 "resolution": "1k"
12 }
13 }'1{
2 "code": 200,
3 "message": "success",
4 "data": {
5 "taskId": "task_1234567890abcdef"
6 }
7}codeStatus code. 200 indicates that the task was created successfully; other values indicate failure.
messageResponse message. For failed requests, this usually contains an error description.
data.taskIdThe ApiPass task ID. Save this value and use it to query task status and results.
flux
flux
Generate photorealistic images with robust text
Starting from
0 credits
Nano Banana 2 Lite, aka Gemini 3.1 Flash-Lite Image, is a newly released image model from Google DeepMind built for fast 1K image generation, efficient creative iteration, prompt-based image editing, and responsive visual output in product-facing creative scenarios.
Starting from
0 credits
OpenAI
OpenAI
Generate intentionally designed visuals with less prompting using the latest OpenAI GPT Image 2 model.
Starting from
0 credits
Qwen
Qwen
Starting from
0 credits
Nano Banana now allows you to freely draw on edited images, bringing you a completely new experience. Simply click on the image to draw whatever you want. You can also add text and details, and Gemini will do the rest.
Starting from
0 credits
Fast image generation model built on Gemini 3.1 Flash Image. Supports text-to-image, image editing, multi-image fusion with up to 14 reference images, Google Search grounding, and outputs up to 4K resolution.
Starting from
0 credits
seedream
seedream
Starting from
0 credits
grok
grok
Grok Imagine is xAI’s AI video generation model for creating high-quality videos with realistic motion, native audio, and synchronized speech. Grok Imagine 1.5 introduces faster generation speeds, improved motion physics, and stronger visual consistency for modern AI video applications.
Starting from
0 credits
qwen
qwen
Generate realistic 2K images with structured text
Starting from
0 credits
seedream
seedream
Starting from
0 credits
wan
wan
Generate images with cinematic precision through high-fidelity text-to-image synthesis and structural editing.
Starting from
0 credits
seedream
seedream
Gemini said Generate images with lightning-fast efficiency and high-fidelity detail through a streamlined, lightweight architecture optimized for instant creative expression.
Starting from
0 creditsqwen
qwen
Generate realistic 2K images with structured text
Starting from
0 credits
Luma
Luma
Starting from
0 credits
seedream
seedream
Starting from
0 credits
Nano Banana 2.1 is Google's latest Nano Banana image model for high-quality text-to-image generation and prompt-based image editing, with up to 10 reference images and 1K, 2K and 4K output.
Starting from
0 credits