Image Generation

Yunxin exposes the OpenAI Images API across every provider that supports image generation. Pick a model with the Models API (GET /v1/models?type=image_generation) and call the same two endpoints regardless of provider.

Which parameters and values a model accepts (sizes, qualities, styles, n range) is per-model. Unsupported options are ignored or rejected upstream — query the model's capabilities via the Models API rather than assuming a fixed set.

Generate images

POST /v1/images/generations

Request

{
  "model": "model-id",
  "prompt": "A serene mountain landscape at sunset, digital art",
  "n": 1,
  "size": "1024x1024",
  "quality": "standard",
  "response_format": "url"
}

Parameters

Parameter	Type	Required	Description
`model`	string	Yes	Image model id.
`prompt`	string	Yes	Text description of the image (up to 4000 chars).
`n`	integer	No	Number of images to generate (1–10). Default `1`.
`size`	string	No	Image size, e.g. `1024x1024`. Default `1024x1024`.
`quality`	string	No	Render quality, e.g. `standard` or `hd` (model-dependent).
`style`	string	No	Style hint, e.g. `vivid` or `natural` (model-dependent).
`response_format`	string	No	`url` (default) or `b64_json`.
`negative_prompt`	string	No	What to avoid (model-dependent).
`seed`	integer	No	Seed for reproducibility (model-dependent).

Some diffusion models also accept width/height, steps, guidance_scale, and strength. These are provider-specific and validated upstream; rely on GET /v1/models/{model_id} for what a given model honors.

Response

{
  "created": 1709251200,
  "data": [
    {
      "url": "https://api.yuhuanstudio.com/v1/temp-images/abc123.png",
      "revised_prompt": "A serene mountain landscape at sunset, digital art"
    }
  ]
}

With response_format: "b64_json", each item carries b64_json instead of url.

Edit images

POST /v1/images/edits

Edit or inpaint an existing image. Send the source image (and an optional mask) as multipart form data along with a prompt:

response = client.images.edit(
    model="model-id",
    image=open("original.png", "rb"),
    mask=open("mask.png", "rb"),
    prompt="Replace the sky with a starry night",
    n=1,
    size="1024x1024",
)

print(response.data[0].url)

The mask is optional; where supported, its transparent pixels mark the region to regenerate.

Examples

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["YUNXIN_API_KEY"],
    base_url="https://api.yuhuanstudio.com/v1",
)

response = client.images.generate(
    model="model-id",
    prompt="A futuristic city with flying cars, cyberpunk style",
    size="1024x1024",
    n=1,
)

print(response.data[0].url)

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.YUNXIN_API_KEY,
  baseURL: "https://api.yuhuanstudio.com/v1",
});

const response = await client.images.generate({
  model: "model-id",
  prompt: "A futuristic city with flying cars, cyberpunk style",
  size: "1024x1024",
  n: 1,
});

console.log(response.data[0].url);

curl https://api.yuhuanstudio.com/v1/images/generations \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $YUNXIN_API_KEY" \
  -d '{
    "model": "model-id",
    "prompt": "A futuristic city with flying cars, cyberpunk style",
    "size": "1024x1024",
    "n": 1
  }'

Image models

List available image models and their capabilities with the Models API: GET /v1/models?type=image_generation. To analyze images as input to a chat model, see Vision instead.

Generate images

POST /v1/images/generations

Request

{
  "model": "model-id",
  "prompt": "A serene mountain landscape at sunset, digital art",
  "n": 1,
  "size": "1024x1024",
  "quality": "standard",
  "response_format": "url"
}

Parameters

Parameter	Type	Required	Description
`model`	string	Yes	Image model id.
`prompt`	string	Yes	Text description of the image (up to 4000 chars).
`n`	integer	No	Number of images to generate (1–10). Default `1`.
`size`	string	No	Image size, e.g. `1024x1024`. Default `1024x1024`.
`quality`	string	No	Render quality, e.g. `standard` or `hd` (model-dependent).
`style`	string	No	Style hint, e.g. `vivid` or `natural` (model-dependent).
`response_format`	string	No	`url` (default) or `b64_json`.
`negative_prompt`	string	No	What to avoid (model-dependent).
`seed`	integer	No	Seed for reproducibility (model-dependent).

Response

{
  "created": 1709251200,
  "data": [
    {
      "url": "https://api.yuhuanstudio.com/v1/temp-images/abc123.png",
      "revised_prompt": "A serene mountain landscape at sunset, digital art"
    }
  ]
}

With response_format: "b64_json", each item carries b64_json instead of url.

Edit images

POST /v1/images/edits

Edit or inpaint an existing image. Send the source image (and an optional mask) as multipart form data along with a prompt:

response = client.images.edit(
    model="model-id",
    image=open("original.png", "rb"),
    mask=open("mask.png", "rb"),
    prompt="Replace the sky with a starry night",
    n=1,
    size="1024x1024",
)

print(response.data[0].url)

The mask is optional; where supported, its transparent pixels mark the region to regenerate.

Examples

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["YUNXIN_API_KEY"],
    base_url="https://api.yuhuanstudio.com/v1",
)

response = client.images.generate(
    model="model-id",
    prompt="A futuristic city with flying cars, cyberpunk style",
    size="1024x1024",
    n=1,
)

print(response.data[0].url)

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.YUNXIN_API_KEY,
  baseURL: "https://api.yuhuanstudio.com/v1",
});

const response = await client.images.generate({
  model: "model-id",
  prompt: "A futuristic city with flying cars, cyberpunk style",
  size: "1024x1024",
  n: 1,
});

console.log(response.data[0].url);

curl https://api.yuhuanstudio.com/v1/images/generations \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $YUNXIN_API_KEY" \
  -d '{
    "model": "model-id",
    "prompt": "A futuristic city with flying cars, cyberpunk style",
    "size": "1024x1024",
    "n": 1
  }'

Image models

List available image models and their capabilities with the Models API: GET /v1/models?type=image_generation. To analyze images as input to a chat model, see Vision instead.

Generate images

Request

Parameters

Response

Edit images

Examples

Image models

On this page

Image Generation

Generate images

Request

Parameters

Response

Edit images

Examples

Image models

On this page