Image Generation

HiAPI supports multiple image generation models through two API formats.

Available Models

Model	Provider	Best For	Credits/Image
Nano Banana	Google	Fast, versatile, 1K-4K	6
Midjourney	Midjourney	Artistic, creative	10
DALL-E 3	OpenAI	Prompt understanding	8
GPT Image 1.5	OpenAI	Text rendering, editing	10
FLUX 1.1 Pro	Black Forest Labs	Photorealism	10

API Formats

OpenAI Images API (/v1/images/generations)

Used by: Midjourney, DALL-E 3, GPT Image 1.5, FLUX 1.1 Pro

Python

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://api.hiapi.ai/v1"
)

response = client.images.generate(
    model="dall-e-3",
    prompt="a futuristic city at sunset",
    size="1024x1024"
)

print(response.data[0].url)

cURL

curl -X POST https://api.hiapi.ai/v1/images/generations \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "dall-e-3",
    "prompt": "a futuristic city at sunset",
    "size": "1024x1024"
  }'

Chat Completions API (/v1/chat/completions)

Used by: Nano Banana (returns image as base64 in message content)

Python

import requests

response = requests.post(
    "https://api.hiapi.ai/v1/chat/completions",
    headers={
        "Authorization": "Bearer YOUR_API_KEY",
        "Content-Type": "application/json"
    },
    json={
        "model": "gemini-3-pro-image-preview",
        "stream": False,
        "messages": [{"role": "user", "content": "a cute cat wearing sunglasses"}],
        "extra_body": {
            "google": {
                "image_config": {"aspect_ratio": "1:1", "image_size": "1K"}
            }
        }
    }
)

# Response contains ![image](data:image/jpeg;base64,...)
print(response.json()["choices"][0]["message"]["content"])

Supported Sizes

Most models support these sizes:

• 1024x1024 (square, default)
• 1792x1024 (landscape)
• 1024x1792 (portrait)

Nano Banana also supports 1024x768 and 768x1024.

Tips

• Use English prompts for best results across all models
• Be specific — “a golden retriever playing fetch on a beach at sunset” works better than “a dog”
• Specify style — add keywords like “watercolor”, “photorealistic”, “cyberpunk” at the end
• Try multiple models — each model has different strengths