Image Generation

Generate stunning images from text prompts using world-class models — GPT-Image-2, Seedream, Imagen 4, Flux, Wan, and more. BelugAPI transparently handles async polling so you always get back image URLs directly.

Transparent async. Image generation is async internally, but BelugAPI polls until the images are ready and returns the URLs directly — you don't need to implement polling yourself.

POST https://api.belugapi.com/v1/images/generations Async (handled)

Requires Authorization: Bearer bapi_… header.

Request parameters

Parameter	Type	Required	Description
model	string	required	Image model ID — see table below.
prompt	string	required	Text description of the image to generate.
n	integer	optional	Number of images to generate. Default: 1. Max: 4 (model-dependent; GPT-Image models: 1).
size	string	optional	Aspect ratio label: `1:1`, `16:9`, `9:16`, `4:3`, `3:4`, `3:2`, `2:3`, `5:4`, `4:5`, `2:1`, `1:2`, `21:9`, `9:21`, `auto`. Default: `1:1`.
quality	string	optional	`standard` or `hd` (model-dependent).
response_format	string	optional	`url` (default) or `b64_json`.

Basic example — GPT-Image-2

from openai import OpenAI

client = OpenAI(
    api_key="bapi_your_key_here",
    base_url="https://api.belugapi.com/v1"
)

result = client.images.generate(
    model="gpt-image-2",
    prompt="A photorealistic beluga whale swimming through crystal clear arctic water, 4K",
    size="16:9",
    n=1,
)

print(result.data[0].url)

import OpenAI from "openai";
const client = new OpenAI({ apiKey: "bapi_your_key_here", baseURL: "https://api.belugapi.com/v1" });

const result = await client.images.generate({
  model:  "gpt-image-2",
  prompt: "A photorealistic beluga whale swimming through crystal clear arctic water, 4K",
  size:   "16:9",
  n:      1,
});

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

curl https://api.belugapi.com/v1/images/generations \
  -H "Authorization: Bearer bapi_your_key_here" \
  -H "Content-Type: application/json" \
  -d '{
    "model":  "gpt-image-2",
    "prompt": "A photorealistic beluga whale swimming through crystal clear arctic water, 4K",
    "size":   "16:9",
    "n": 1
  }'

Response

200 OK

{
  "created": 1716900000,
  "data": [
    {
      "url": "https://cdn.example.com/generated/image-abc123.png"
    }
  ]
}

Seedream (ByteDance)

result = client.images.generate(
    model="seedream-4.5",
    prompt="Anime-style warrior princess in a cherry blossom garden, vibrant colors",
    size="9:16",
)

curl https://api.belugapi.com/v1/images/generations \
  -H "Authorization: Bearer bapi_your_key_here" \
  -H "Content-Type: application/json" \
  -d '{
    "model":  "seedream-4.5",
    "prompt": "Anime-style warrior princess in a cherry blossom garden, vibrant colors",
    "size":   "9:16"
  }'

Imagen 4.0 (Google)

result = client.images.generate(
    model="gpt-image-2",
    prompt="Hyper-realistic product photo of a glass bottle of mineral water, studio lighting",
    size="1:1",
)

curl https://api.belugapi.com/v1/images/generations \
  -H "Authorization: Bearer bapi_your_key_here" \
  -H "Content-Type: application/json" \
  -d '{
    "model":  "gpt-image-2",
    "prompt": "Hyper-realistic product photo of a glass bottle of mineral water, studio lighting",
    "size":   "1:1"
  }'

Flux 2.0

result = client.images.generate(
    model="flux-2.0",
    prompt="Oil painting of a Renaissance-style portrait of a tech entrepreneur",
    size="3:4",
)

curl https://api.belugapi.com/v1/images/generations \
  -H "Authorization: Bearer bapi_your_key_here" \
  -H "Content-Type: application/json" \
  -d '{
    "model":  "flux-2.0",
    "prompt": "Oil painting of a Renaissance-style portrait of a tech entrepreneur",
    "size":   "3:4"
  }'

Wan 2.7 Image

result = client.images.generate(
    model="wan2.7-image-pro",
    prompt="Futuristic city skyline at night with neon reflections on rain-soaked streets",
    size="21:9",
)

curl https://api.belugapi.com/v1/images/generations \
  -H "Authorization: Bearer bapi_your_key_here" \
  -H "Content-Type: application/json" \
  -d '{
    "model":  "wan2.7-image-pro",
    "prompt": "Futuristic city skyline at night with neon reflections on rain-soaked streets",
    "size":   "21:9"
  }'

Available image models

Model ID	Name	Provider	Aspect ratios / sizes
`gpt-image-2`	GPT-Image-2	OpenAI	auto, 1:1, 16:9, 9:16, 4:3 …
`gpt-image-2-official`	GPT-Image-2 (Official Channel)	OpenAI	auto, 1:1, 16:9, 9:16, 4:3 …
`gpt-image-1-official`	GPT-Image-1	OpenAI	1:1, 16:9, 9:16, 4:3, 3:4
`gpt-image-1.5-official`	GPT-Image-1.5	OpenAI	1:1, 16:9, 9:16, 4:3, 3:4
`gemini-3-pro-image-preview-official`	Gemini 3 Pro Image (Nano Banana Pro)	Google	1:1, 16:9, 9:16, 4:3, 3:4
`gemini-2.5-flash-image-preview-official`	Gemini 2.5 Flash Image (Nano Banana)	Google	1:1, 16:9, 9:16, 4:3, 3:4
`seedream-4.0`	Seedream 4.0	ByteDance	1:1, 16:9, 9:16, 4:3, 3:4
`seedream-4.5`	Seedream 4.5	ByteDance	1:1, 16:9, 9:16, 4:3, 3:4
`seedream-5.0-lite`	Seedream 5.0 Lite	ByteDance	1:1, 16:9, 9:16, 4:3, 3:4
`wan2.7-image`	Wan 2.7 Image	Various	—
`wan2.7-image-pro`	Wan 2.7 Image Pro	Various	—
`qwen-image-2.0-pro`	Qwen Image 2.0 Pro	Various	—
`z-image-turbo`	Z-Image-Turbo	Various	—
`grok-imagine-1.0-apimart`	Grok Imagine 1.0	Various	—
`grok-imagine-1.0-edit-apimart`	Grok Imagine 1.0 Edit	Various	—
`flux-kontext`	Flux Kontext	Various	—
`flux-2.0`	Flux 2.0	Various	—

Size / aspect ratio reference

Use these values in the size field. Pixel-style values like 1024x1024 are accepted for compatibility but are automatically converted.

Value	Description	Best for
1:1	Square	Social media, avatars, icons
16:9	Landscape widescreen	Thumbnails, banners, wallpapers
9:16	Portrait (mobile)	Stories, Reels, Shorts
4:3	Classic landscape	Presentations, print
3:4	Portrait	Photos, posters
21:9	Ultra-wide	Cinematic headers
3:2	DSLR landscape	Photography-style
auto	Model decides	Let the AI choose