Anbu AI Documentation

Anbu AI is an open-source platform for building AI applications. Access powerful language models and image generation APIs with simple REST endpoints.

No API Key RequiredOpen Source

Quick Start

Get started with a simple API call. No authentication required.

Chat Completion Request

bash

curl -X POST /api/chat \
  -H "Content-Type: application/json" \
  -H "X-Session-Id: my-session" \
  -d '{
    "message": "Hello, how are you?",
    "model": "gpt-4o"
  }'

Response

json

{
  "text": "Hello! I'm doing well, thank you for asking. How can I assist you today?",
  "model": "gpt-4o",
  "citations": []
}

Chat Completions

POST/api/chat

Generate text completions using various AI models. Supports conversation history via session management.

Request Headers

Header	Required	Description
Content-Type	Yes	application/json
X-Session-Id	No	Unique session identifier for conversation history

Request Body

Parameter	Type	Required	Description
message	string	Yes	The user message to send
model	string	No	Model to use: gpt-4o, gpt-3.5 (default: gpt-4o)
systemPrompt	string	No	Custom system instructions for the model

Example Request

javascript

const response = await fetch('/api/chat', {
  method: 'POST',
  headers: {
    'Content-Type': 'application/json',
    'X-Session-Id': 'user-123'
  },
  body: JSON.stringify({
    message: 'Explain quantum computing in simple terms',
    model: 'gpt-4o',
    systemPrompt: 'You are a helpful science tutor'
  })
});

const data = await response.json();
console.log(data.text);

Image Generation

POST/api/image

Generate images using AI models with various styles and sizes.

Request Body

Parameter	Type	Required	Description
prompt	string	Yes	Description of the image to generate
style	string	No	Style: default, ghibli, cyberpunk, anime, portrait, 3d
size	string	No	Aspect ratio: 1:1, 3:2, 2:3 (default: 1:1)

Example Request

javascript

const response = await fetch('/api/image', {
  method: 'POST',
  headers: { 'Content-Type': 'application/json' },
  body: JSON.stringify({
    prompt: 'A futuristic city at sunset with flying cars',
    style: 'cyberpunk',
    size: '1:1'
  })
});

const data = await response.json();
console.log(data.imageUrl);

Response

json

{
  "success": true,
  "model": "deepimg",
  "imageUrl": "https://example.com/generated-image.png",
  "prompt": "A futuristic city at sunset with flying cars"
}

Session Management

POST/api/reset

Reset conversation history for a specific session.

Reset Conversation

bash

curl -X POST /api/reset \
  -H "X-Session-Id: my-session"

Response

json

{
  "status": "ok",
  "message": "Conversation reset",
  "sessionId": "my-session",
  "hadHistory": true
}

Status

GET/api/status

Check the API status and available models.

Response

json

{
  "status": "ok",
  "version": "1.0.0",
  "models": ["gpt-4o", "gpt-3.5", "deepimg"]
}

GPT-4o Model

GPT-4o is the most capable model available, offering advanced reasoning, longer context, and vision capabilities.

Parameters

Parameter	Default	Description
temperature	0.9	Controls randomness (0-1)
max_tokens	2048	Maximum response length
top_p	0.7	Nucleus sampling parameter

GPT-3.5 Model

GPT-3.5 is a fast and efficient model ideal for simpler tasks and quick responses.

DeepImg Model

DeepImg is a high-quality image generation model supporting multiple artistic styles.

Available Styles

default

ghibli

cyberpunk

anime

portrait

3d