Generate image assets with steadier cost control
Move product shots, covers, campaign drafts, and repeated image jobs into a dedicated image lane. See what each size is likely to cost, which key was used, and where the records live.
Supported models
GPT Image 2.5 Flare and GPT Image 2.5 Sunburst are now available at the same per-image price as GPT Image 2.
Generate image assets with steadier cost control
Image generation behaves differently from chat: jobs can take longer, retries happen, sizes matter, and cost checks matter. Stable Images separates that work so business users can understand spend and developers can connect or debug it cleanly.
Image jobs, keys, and records stay separate from normal chat and text usage.
Per-size estimates help you plan a batch before you generate it.
Developers can use an OpenAI Images-compatible request shape instead of learning a custom flow.
Not a one-off prompt toy. A lane for image production.
Use Stable Images when you need repeatable output, budget visibility, and a setup your team or system can keep using.
More product image options
Generate product shots, white-background images, lifestyle visuals, and size variants so your team can choose from useful options faster.
Campaign assets with less waiting
Turn article images, social covers, poster drafts, and brand assets into a repeatable workflow instead of starting from scratch every time.
Batch jobs can run automatically
Connect a product backend or script to generate many SKU images, cover variants, or template-based assets without manual clicking.
GPT Image 2 / GPT Image 2.5
Stable Image API charges each generated image once by its final size: $1.5 for 1K, $3 for 2K, and $4 for 4K. The table classifies a size by its longest edge; RMB ranges are estimated from the current buying power of balance and subscription plans.
| Size | Billing boundary | Charge per image | Estimated RMB per image |
|---|---|---|---|
| 1K | Longest edge <= 1024px | $1.5 / image | ¥0.23 - ¥0.60 / image |
| 2K | Longest edge 1025-2048px | $3 / image | ¥0.45 - ¥1.20 / image |
| 4K | Longest edge > 2048px | $4 / image | ¥0.60 - ¥1.60 / image |
Billing uses the actual output size returned by the upstream service first, then falls back to the requested size. auto, empty, or unrecognizable sizes fall back to 2K.
https://api.clawto.link/image/v1
If you own the business need, start by checking the use case, pricing, and budget. When it is time to connect a product or script, send the key and setup details to your developer.
Sign in to GetToken and open Stable Image API in the console.
Create an image-only key and name it after the project or workflow.
Send the key, endpoint, and model parameters to the product, script, or developer using it.
Review balance usage, key status, and records in the console after generation starts.
Most users do not need to understand every parameter. This section keeps the connection details in one place for whoever integrates the workflow.
POST https://api.clawto.link/image/v1/images/generations
Authorization: Bearer <image_api_key>
{
"model": "gpt-image-2.5-flare",
"prompt": "A clean product photo on a white desk",
"size": "1024x1024",
"quality": "auto",
"output_format": "png"
}FAQ
Turn image generation into a managed workflow
Create an image key, connect product visuals or campaign assets, and scale usage after you understand per-image cost.
Continue with GetToken
Compare plans, follow the setup guide, or check product and support details.