legnext 0.1.3

Python SDK for Legnext AI API - Professional image and video generation using Midjourney models

Legnext Python SDK

Official Python client library for the Legnext AI API - Professional image and video generation using Midjourney models.

Features

• Comprehensive API Coverage: Support for all Legnext endpoints
• Type Safety: Full type hints with Pydantic models
• Async Support: Both synchronous and asynchronous clients
• Modern Design: Clean, intuitive API following OpenAI SDK patterns
• Task Management: Built-in polling and webhook support
• Error Handling: Robust error handling with automatic retries

Installation

pip install legnext

Or with uv:

uv pip install legnext

Quick Start

Basic Usage

from legnext import Client

client = Client(api_key="your-api-key")

# Generate an image
response = client.midjourney.diffusion(
    text="a beautiful sunset over mountains"
)

print(f"Job ID: {response.job_id}")
print(f"Status: {response.status}")

# Wait for completion
result = client.tasks.wait_for_completion(response.job_id)
print(f"Image URLs: {result.output.image_urls}")

Async Usage

import asyncio
from legnext import AsyncClient

async def main():
    async with AsyncClient(api_key="your-api-key") as client:
        response = await client.midjourney.diffusion(
            text="a futuristic cityscape"
        )
        result = await client.tasks.wait_for_completion(response.job_id)
        print(f"Generated: {result.output.image_urls}")

asyncio.run(main())

API Coverage

Midjourney Operations

Image Generation:

• client.midjourney.diffusion(text) - Text to image generation (POST /diffusion)
• client.midjourney.variation(job_id, image_no, type) - Create variations (POST /variation)
• client.midjourney.upscale(job_id, image_no, type) - Upscale images (POST /upscale)
• client.midjourney.reroll(job_id) - Re-generate with same prompt (POST /reroll)
• client.midjourney.blend(image_urls) - Blend 2-5 images (POST /blend)
• client.midjourney.describe(image_url) - Generate descriptions (POST /describe)
• client.midjourney.shorten(prompt) - Optimize prompts (POST /shorten)
• client.midjourney.pan(job_id, image_no, direction) - Extend in direction (POST /pan)
• client.midjourney.outpaint(job_id, image_no) - Expand all directions (POST /outpaint)
• client.midjourney.inpaint(job_id, image_no, mask, prompt) - Region editing (POST /inpaint)
• client.midjourney.remix(job_id, image_no, prompt) - Transform with new prompt (POST /remix)
• client.midjourney.edit(job_id, image_no, prompt) - Edit specific areas (POST /edit)
• client.midjourney.upload_paint(image, prompt) - Advanced editing (POST /upload-paint)
• client.midjourney.retexture(job_id, image_no, prompt) - Change textures (POST /retexture)
• client.midjourney.remove_background(job_id, image_no) - Remove background (POST /remove-background)
• client.midjourney.enhance(job_id, image_no) - Improve quality (POST /enhance)

Video Generation:

• client.midjourney.video_diffusion(prompt, image_url, duration) - Generate video (POST /video-diffusion)
• client.midjourney.extend_video(job_id) - Extend video (POST /extend-video)
• client.midjourney.video_upscale(job_id) - Upscale video (POST /video-upscale)

Task Management

• client.tasks.get() - Get task status
• client.tasks.wait_for_completion() - Poll until complete

Documentation

• Full API Documentation
• Examples
• Official API Reference

Development

Setup

# Clone the repository
git clone https://github.com/your-org/legnext-python
cd legnext-python

# Install with dev dependencies
uv pip install -e ".[dev]"

Running Tests

pytest

Code Quality

# Format code
black src tests
isort src tests

# Lint
ruff check src tests

# Type check
mypy src

Requirements

• Python 3.10+
• httpx
• pydantic

License

Apache License 2.0

Support

• Email: [e-mail]
• Documentation: https://legnext.ai/docs
• Issues: https://github.com/your-org/legnext-python/issues

Changelog

See CHANGELOG.md for version history.