Documentation

System Requirements

• Python 3.8 or higher
• Linux (Ubuntu 20.04+, Fedora 35+, Arch Linux)
• 4GB RAM minimum (8GB recommended)
• 500MB free storage
• Ollama (optional, for local models)

Quick Installation

git clone https://github.com/AutoBotSolutions/AI-Chat-Linux-Client.git
cd chat-linux-client
./scripts/install.sh
./scripts/run.sh

Manual Installation

python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt
python main.py

Installing Ollama (Optional)

For offline AI support with local models:

curl -fsSL https://ollama.ai/install.sh | sh
ollama pull llama3.2:1b
ollama pull mistral:7b

API Keys

Add API keys through the application settings or environment variables:

GROQ_API_KEY=gsk_your_key_here
HUGGINGFACE_API_KEY=hf_your_key_here
OPENROUTER_API_KEY=sk-or-your_key_here
OPENAI_API_KEY=sk-your_key_here

Configuration File

Configuration is stored at ~/.config/chat-linux-client/config.json:

{
  "providers": {
    "groq": {
      "enabled": true,
      "api_key": "your_api_key",
      "base_url": "https://api.groq.com/openai/v1"
    },
    "ollama": {
      "enabled": true,
      "base_url": "http://localhost:11434"
    }
  },
  "chat": {
    "temperature": 0.7,
    "max_tokens": null,
    "routing_strategy": "offline_first"
  }
}

Basic Chat

1. Launch the application
2. Select a model from the dropdown
3. Type your message
4. Press Enter or click Send

Model Selection

Choose models based on your needs:

• For speed: Groq or lightweight Ollama models
• For quality: GPT-4 or capable Ollama models
• For privacy: Ollama local models
• For cost: Free tier providers or local models

Keyboard Shortcuts

Ollama (Local)

Run AI models locally with complete privacy. No API costs, works offline.

Available Models: Llama3.2, Qwen2.5, Phi3.5, Mistral, CodeLlama

Groq

Ultra-low latency inference using LPU technology. Generous free tier.

Available Models: Llama3-8B, Mixtral-8x7B, Gemma-7B

HuggingFace

Access thousands of open-source models. Free tier available.

Available Models: Qwen2.5, Mistral-7B, DeepSeek-R1, DialoGPT

OpenRouter

Access multiple models from various providers through one API.

Available Models: GPT-4, Claude-3, GPT-3.5

OpenAI

State-of-the-art GPT models with excellent quality.

Available Models: GPT-4o, GPT-4-Turbo, GPT-3.5

Prerequisites

• Python 3.8 or higher
• Git
• Virtual environment

Setup

git clone https://github.com/AutoBotSolutions/AI-Chat-Linux-Client.git
cd chat-linux-client
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt
pip install pytest pytest-qt black flake8 mypy

Running Tests

pytest tests/
pytest --cov=. tests/
pytest -v tests/

Test Structure

Tests are organized by module in the tests/ directory:

• test_api_client.py - API client tests
• test_provider_router.py - Routing tests
• test_key_handler.py - Key storage tests
• test_main_window.py - UI tests

Writing Tests

Use pytest with fixtures for clean test code:

import pytest
from core.groq_client import GroqClient

@pytest.fixture
def client():
    return GroqClient(api_key="test_key")

def test_initialization(client):
    assert client.api_key == "test_key"

Layered Architecture

• UI Layer: PyQt6 desktop interface
• Routing Layer: Intelligent model selection
• Provider Layer: AI provider implementations
• Storage Layer: Configuration and history
• Utility Layer: Helper functions

Key Components

• core/ - AI provider logic
• ui/ - User interface
• storage/ - Data persistence
• utils/ - Utility modules

Contribution Guidelines

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Add tests
5. Ensure all tests pass
6. Submit a pull request

Code Style

We follow standard Python conventions:

black .
flake8 .
mypy .

Security Features

• No telemetry or analytics
• Encrypted API key storage
• Optional chat encryption
• HTTPS-only communications
• Local-first data storage

Reporting Vulnerabilities

For security issues, email: security@example.com

Do NOT open public issues for security vulnerabilities.

Application Won't Start

Run system checks:

python main.py --check-system

API Key Not Working

• Verify the key is correct
• Check provider account is active
• Ensure key has proper permissions
• Try regenerating the key

Models Not Showing

• Verify provider is enabled
• Check API key is configured
• Ensure Ollama is running (for local models)

Is Chat Linux Client free?

Yes! It's open source and free. Local models are completely free. Some cloud providers have free tiers.

Can I use it offline?

Yes! With Ollama installed, you can use local models entirely offline.

Which provider should I use?

• For privacy: Ollama (local)
• For speed: Groq
• For quality: OpenAI
• For cost: Ollama or free tiers