Skip to main content

​
Overview

WaveAPI supports multiple image generation models, all accessible through the unified async task endpoint /v1/tasks.
Simple synchronous image generation can also use /v1/tasks (OpenAI compatible format), but /v1/tasks is recommended for broader model and parameter support.

​
Request Endpoint

POST https://qingbo.dev/v1/tasks

​
Common Request Parameters

ParameterTypeRequiredDescription
modelstringYesModel name
actionstringNoOperation type, defaults to "generate", options include "edit", "upscale", "variation"
promptstringYesImage description text
negative_promptstringNoContent you do not want to appear
nintegerNoNumber of images to generate, default 1
sizestringNoImage size, e.g., "1024x1024"
aspect_ratiostringNoAspect ratio, e.g., "16:9"
qualitystringNoQuality level, e.g., "standard", "hd"
image_urlstringNoReference image URL (image-to-image)
image_urlsarrayNoMultiple reference image URLs
callback_urlstringNoWebhook callback URL upon completion
extraobjectNoModel-specific pass-through parameters

​
Response Format

After task submission, a task_id is returned. Query results via GET /v1/tasks/{task_id}: 
{
  "task_id": "task_xxx",
  "status": "completed",
  "result": {
    "images": [
      {
        "url": "https://...",
        "expires_at": 1720000000
      }
    ]
  }
}

​
Supported Models

GPT-4o Image

OpenAI image generation model

FLUX Kontext

FLUX series image generation

Seedream

ByteDance Seedream series

More Models

Nano Banana, Gemini Flash, and more
Each model supports different parameters. The extra field is used to pass model-specific parameters. Refer to each model’s documentation page for detailed parameter support.