Skip to main content
Generate images from text prompts, edit existing images with natural language, or transform images with style transfer. All generated images are automatically saved to your Space.

​
Quick Start

import requests

response = requests.post(
    "https://api.zarkai.xyz/v1/chat",
    headers={
        "x-api-key": "your_api_key",
        "Content-Type": "application/json",
    },
    json={
        "messages": [
            {"role": "user", "content": "Create an image of a futuristic cityscape at night with neon lights"}
        ],
        "workspace_id": "wks-xxxxx",
        "tools": ["image_gen"],
        "tool_choice": "auto",
    },
)

data = response.json()
print(data["response"])
Generated images are returned as file references and saved directly to your Space.

​
Image Editing

Edit an existing image by referencing a file already in your Space: 
{
  "messages": [
    {"role": "user", "content": "Add falling snow to this image"}
  ],
  "workspace_id": "wks-xxxxx",
  "tools": ["image_gen"],
  "file_ids": ["file-abc123"],
  "tool_choice": "auto"
}
The model understands the image content and applies your requested changes while preserving the rest of the scene.

​
Style Transfer

Transform existing images by describing the desired aesthetic: 
{
  "messages": [
    {"role": "user", "content": "Render this image as an oil painting in the style of impressionism"}
  ],
  "workspace_id": "wks-xxxxx",
  "tools": ["image_gen"],
  "file_ids": ["file-abc123"],
  "tool_choice": "auto"
}
Supported styles include oil painting, watercolor, pencil sketch, pop art, anime, pixel art, and any style you can describe.

​
Streaming

Python
import requests
import json

response = requests.post(
    "https://api.zarkai.xyz/v1/chat",
    headers={"x-api-key": "your_api_key", "Content-Type": "application/json"},
    json={
        "messages": [{"role": "user", "content": "Create an image of a mountain landscape at sunrise"}],
        "workspace_id": "wks-xxxxx",
        "tools": ["image_gen"],
        "stream": True,
    },
    stream=True,
)

for line in response.iter_lines():
    if line:
        event = json.loads(line.decode("utf-8").removeprefix("data: "))
        if event["type"] == "media_item":
            print(f"Image generated: {event['file_id']}")
        elif event["type"] == "ai_chunk":
            print(event["content"], end="", flush=True)
        elif event["type"] == "ai_complete":
            print(f"\n\nDone.")

​
Streaming Events

EventDescription
media_itemGenerated image file reference with file_id and metadata
ai_chunkIncremental text describing what was generated
ai_completeFinal response with image reference

​
What Image Generator Is Not

Image Generator is a creation tool. It generates new visual content from descriptions.

​
Examples

Create an image of a futuristic cityscape at night with neon lights

Generate a minimalist logo for a coffee shop called "Brew"

Draw a technical diagram showing a load balancer architecture

Render this photo as a pencil sketch

Create an icon set for a mobile app — home, search, profile, settings

Apply a watercolor style to my uploaded photo

Edit this image to add rain