AlleAI Python SDK

A Python SDK for interacting with the AlleAI platform, providing seamless access to state-of-the-art AI models for image generation, audio processing, and video creation.

Features

• Chat Completions API – Generate conversational responses using multiple AI models such as GPT-5 and Claude Sonnet 4.6. Optional tools and tool follow-up via request_log_id + tool_result.
• Embeddings API – Create vectors with cohere-embed-v4. Alle-AI does not store embeddings.
• Tools API – Execute built-in file tools (generate_pdf, generate_docx, generate_pptx, generate_xlsx, generate_chart).
• Image Generation & Editing API – Create and manipulate high-quality images using various AI models like Grok-2 and DALL·E 3.
• Audio API – Multi-model support for:
  • Text-to-Speech (TTS): Convert text into natural-sounding audio.
  • Speech-to-Text (STT): Transcribe spoken audio into text.
  • Audio Generation: Create unique audio content using generative AI.
• Video Generation API – Generate AI-powered videos with customizable parameters, backed by multiple AI models.

Installation

Install the package using pip:

pip install alle-ai-sdk

Quick Start

Environment Setup

Create a .env file in your project root:

ALLEAI_API_KEY=your_api_key_here

Initialize Client

from alleai.core import AlleAIClient
from dotenv import load_dotenv
import os

# Load environment variables
load_dotenv()

# Get API key
api_key = os.getenv("ALLEAI_API_KEY")

# Initialize client
client = AlleAIClient(api_key=api_key)

Usage Examples

Chat Completions

# Chat completion example
chat = client.chat.completions({
    "models": ["gpt-5", "claude-sonnet-4-6"],
    "messages": [
        {
            "system": [
                {
                    "type": "text",
                    "text": "You are a helpful assistant."
                }
            ]
        },
        {
            "user": [
                {
                    "type": "text",
                    "text": "tell me about photosynthesis?"
                }
            ]
        }
    ],
    "temperature": 0.7,      # Controls randomness (0.0 to 1.0)
    "max_tokens": 2000,      # Maximum response length
    "top_p": 1,              # Controls diversity
    "frequency_penalty": 0.2, # Penalizes repeated tokens
    "presence_penalty": 0.3,  # Penalizes new tokens
    "stream": False          # Enable streaming responses
})
print(chat)

Image Generation

# Generate an image
image_response = client.image.generate({
    "models": ["grok-2-image", "dall-e-3"],
    "prompt": "futuristic city, flying cars, robotic pedestrians.",
    "model_specific_params": {
        "grok-2-image": {
            "n": 1,
            "height": 1024,
            "width": 1024
        }
    },
    "n": 1,
    "height": 1024,
    "width": 1024,
    "seed": 8  # For reproducible results
})
print(image_response)

Image Editing

# Edit an existing image
response = client.image.edit({
    "models": ["nova-canvas"],
    "image_file": "path/to/image.jpg",
    "prompt": "Replace the sky with a sunset"
})
print(response)

Text-to-Speech

# Convert text to speech
tts_response = client.audio.tts({
    "models": ["elevenlabs-multilingual-v2", "gpt-4o-mini-tts"],
    "prompt": "Hello! You're listening to a test of a text-to-speech model...",
    "voice": "nova",
    "model_specific_params": {
        "gpt-4o-mini-tts": {
            "voice": "alternative-voice"
        }
    }
})
print(tts_response)

Speech-to-Text

# Transcribe audio
stt_response = client.audio.stt({
    "models": ["whisper-v3"],
    "audio_file": "path/to/your/audio.mp3"
})
print(stt_response)

Audio Generation

# Generate Audio
response = client.audio.generate({
    "models": ["lyria"],
    "prompt": "Create a relaxing ambient track"
})
print(response)

Video Generation

# Generate a video
video = client.video.generate({
    "models": ["nova-reel"],
    "prompt": "robotic arm assembling a car in a futuristic factory",
    "duration": 6,           # Video length in seconds
    "loop": False,           # Enable looping
    "aspect_ratio": "16:9",  # Video aspect ratio
    "fps": 24,              # Frames per second
    "dimension": "1280x720", # Video dimensions
    "resolution": "720p",    # Output resolution
    "seed": 8               # For reproducible results
})
print(video)

Video Status Check

# Check video generation status
response = client.video.get_video_status({
    "requestId": "your-request-id"
})
print(response)

Embeddings

vectors = client.embeddings.create({
    "input": ["Alle-AI embeddings", "second document"],
    "model": "cohere-embed-v4",
    "input_type": "search_document",
})
print(vectors)

Tools

pdf = client.tools.execute({
    "tool_name": "generate_pdf",
    "arguments": {
        "filename": "report",
        "title": "Q3 Report",
        "content": [{"type": "heading", "text": "Summary", "level": 1}],
    },
})
print(pdf)

Error Handling

from alleai.core.exceptions import InvalidRequestError

try:
    response = client.image.generate({
        "models": ["nova-canvas"],
        "prompt": "Your prompt here"
    })
except InvalidRequestError as e:
    print(f"Error: {e.message}")
    print(f"Error Code: {e.code}")

Requirements

• Python 3.8 or higher
• Valid AlleAI API key
• requests library (automatically installed with the package)

Support

For technical support and inquiries:

• Open an issue in our GitHub repository
• Contact our support team at contact@alle-ai.com

License

This project is licensed under the MIT License - see the LICENSE file for details.

Security