Inputs
A text prompt; models explicitly tagged for image-to-image or editing may also accept a source image.
image capability center
Image API: generation and editing models
Choose Chinese image models by verified text-to-image, image-to-image, editing, and resolution capabilities, then run the correct API request.
Contract
Inputs, outputs, and lifecycle.
Outputs
A generated or edited image payload/URL in the format supported by the selected model.
Endpoint
POST /v1/images/generations
Lifecycle
Synchronous request. Save the returned image URL or base64 payload before it expires.
Supported tasks
One definition per task.
text-to-image, image-to-image, image-editing, high-resolution
Limit: Do not assume editing, reference images, 4K output, aspect ratios, or batch sizes unless the selected model lists that capability.
Matching catalog
6 models with explicit metadata.
Unknown capabilities are omitted, never guessed from a model name. Pricing examples use the synchronized public data; live Dash Pricing remains authoritative.
step-2x-large
StepFun Step 2X Large — text-to-image generation with realistic texture and strong Chinese and English text rendering inside the image itself. Supports 256x256, 512x512, 768x768, 1024x1024, 1280x800, and 800x1280; one image per request over the OpenAI-compatible images endpoint.
Tasks: text-to-image
Endpoint: POST /v1/images/generations
Cost example: 1 image × $0.0155 = $0.0155.
step-image-edit-2
StepFun Step Image Edit 2 — unified text-to-image and image-editing model under 6B parameters that completes an edit in one to two seconds, built for low-latency interactive retouching. Supports negative prompts and a text-optimization mode; one image per request over the OpenAI-compatible images endpoint.
Tasks: image-editing
Endpoint: POST /v1/images/generations
Cost example: 1 image × $0.0035 = $0.0035.
image-01
MiniMax image-01 — text-to-image generation over MiniMax's native image_generation endpoint, billed per generated image. The vendor lists it under legacy models.
Tasks: text-to-image
Endpoint: POST /v1/images/generations
Cost example: 1 image × $0.0035 = $0.0035.
doubao-seedream-5-0-260128
ByteDance Doubao Seedream 5.0-lite — text-to-image and image-to-image with smarter controllable creation, real-time retrieval, and stronger subject consistency (2K+ resolution).
Tasks: text-to-image, image-to-image, high-resolution
Endpoint: POST /v1/images/generations
Cost example: 1 image × $0.030556 = $0.030556.
wan2.7-image
Alibaba Wan2.7 Image — text-to-image, image editing, multi-image reference generation, interactive editing, strong text rendering and instruction following.
Tasks: text-to-image, image-editing
Endpoint: POST /v1/images/generations
Cost example: 1 image × $0.027778 = $0.027778.
wan2.7-image-pro
Alibaba Wan2.7 Image Pro — text-to-image and editing with brand-color control, character-consistent multi-image generation, and output up to 4096×4096.
Tasks: text-to-image, image-editing, high-resolution
Endpoint: POST /v1/images/generations
Cost example: 1 image × $0.069444 = $0.069444.
Run or estimate before registration.
Calculate API cost · Generate curl, Python, or JavaScript · Install the Seedance 2 Agent Recipe
Source and method: public model pricing dataset, updated 2026-08-08. Limitations are stated above; third-party claims remain next to the model pages that cite them.