Installation¶

This guide covers different ways to install the InvokeAI Python Client.

Installation Methods¶

Using Pixi (Recommended)¶

Pixi provides the best development experience with automatic dependency management:

# Install pixi if you haven't already
curl -fsSL https://pixi.sh/install.sh | bash

# Add to your project
pixi add invokeai-py-client

# Or clone and develop
git clone https://github.com/CodeGandee/invokeai-py-client
cd invokeai-py-client
pixi run dev-setup

Using pip¶

Install from PyPI:

pip install invokeai-py-client

Or install the latest development version:

pip install git+https://github.com/CodeGandee/invokeai-py-client.git

Development Installation¶

For contributing or testing the latest features:

# Clone the repository
git clone https://github.com/CodeGandee/invokeai-py-client
cd invokeai-py-client

# Using pixi (recommended)
pixi run dev-setup

# Or using pip
pip install -e ".[dev]"

Dependencies¶

The client installs these core dependencies:

requests — HTTP client for REST API calls
python-socketio — Socket.IO client for real-time events
pydantic — Data validation and models
Pillow — Image operations (used in examples)
rich — Terminal formatting (used in examples)

Note: - JSONPath libraries are not required; the client performs value substitution without JSONPath.

Verifying Installation¶

Test your installation with this simple script:

from invokeai_py_client import InvokeAIClient

# Connect to your InvokeAI server
client = InvokeAIClient.from_url("http://localhost:9090")

# Quick probe (True/False)
print("Health:", client.health_check())

# List boards (requires a running server)
try:
    boards = client.board_repo.list_boards()
    print(f"Boards available: {len(boards)}")
except Exception as e:
    print(f"Failed to list boards (is the server running?): {e}")

Environment Setup¶

InvokeAI Server¶

Ensure your InvokeAI server is running:

# Default server location
http://localhost:9090

# Check if accessible
curl http://localhost:9090/api/v1/app/version

Python Environment¶

We recommend using a virtual environment:

venvcondapixi

python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate
pip install invokeai-py-client

conda create -n invokeai python=3.10
conda activate invokeai
pip install invokeai-py-client

pixi init
pixi add python>=3.9
pixi add invokeai-py-client

Platform-Specific Notes¶

Windows¶

Use PowerShell or Windows Terminal for best results
Path separators: Use raw strings (r"C:\path\to\file") or forward slashes

macOS¶

Install Xcode Command Line Tools if prompted
Use Homebrew to install Python 3.9+ if needed

Linux¶

Most distributions include Python 3.9+
Install python3-dev package if you encounter build errors

Troubleshooting¶

Connection Refused¶

If you get a connection error:

Check if InvokeAI is running: curl http://localhost:9090/api/v1/app/version
Verify the correct URL and port
Check firewall settings

Import Errors¶

If imports fail:

Verify installation: pip show invokeai-py-client
Check Python version: python --version (must be 3.9+)
Reinstall: pip install --force-reinstall invokeai-py-client

SSL/TLS Errors¶

For self-signed certificates:

import urllib3
urllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning)

client = InvokeAIClient.from_url("https://localhost:9090", verify_ssl=False)

Security Note

Only disable SSL verification for local development or trusted servers.

Next Steps¶

With the client installed, proceed to the Quick Start guide to run your first workflow!