Documentation

Complete guide to using Xibe Chat. Learn how to get started, configure models, and leverage advanced features.

Installation

Download from GitHub

Visit the Releases page and download the appropriate version for your platform.

Windows

Download .exe or .zip

Android

Download .apk file

Linux

Download .deb or .AppImage

macOS

Download .dmg file

Quick Start

1. Configure API Keys

Navigate to Settings → AI Models and add your API keys:

1

Open Settings

Click the gear icon in the top right

2

Add API Keys

Enter keys for OpenAI, Anthropic, Google, etc.

3

Select Default Model

Choose your preferred model

2. Start Chatting

Click "New Chat" and start your conversation. The app automatically:

  • Streams responses in real-time
  • Saves conversations locally
  • Maintains conversation history

Configuration

API Keys Configuration

Configure API keys for different providers:

OpenAI

Get your API key from platform.openai.com

Settings → AI Models → OpenAI → Add API Key

Anthropic Claude

Get your API key from console.anthropic.com

Settings → AI Models → Anthropic → Add API Key

Google Gemini

Get your API key from Google AI Studio

Settings → AI Models → Google → Add API Key

AI Models

Xibe Chat supports multiple AI providers. Each model has unique strengths:

GPT-4 Turbo

Best for: Complex reasoning, creative writing, detailed analysis

  • • 128K context window
  • • Vision capabilities
  • • Function calling

Claude 3 Opus

Best for: Long documents, nuanced conversations, coding

  • • 200K context window
  • • Enhanced safety
  • • Code generation

Gemini 1.5 Pro

Best for: Multimodal tasks, large contexts, fast responses

  • • 1M+ context window
  • • Video understanding
  • • Efficient processing

Local Models (Ollama)

Best for: Privacy, offline usage, no API costs

  • • 100% private
  • • No internet needed
  • • Free to use

AI Profiles

Profiles let you customize AI behavior with system prompts. They're completely optional!

Creating a Profile

  1. 1Navigate to Settings → AI Profiles
  2. 2Click "Create New Profile"
  3. 3Give it a name and description
  4. 4Write your system prompt (how the AI should behave)

Example: Code Assistant Profile

System Prompt:

You are an expert programmer. Provide concise, well-commented code examples. Focus on best practices, clean architecture, and performance. Explain your reasoning when helpful.

Privacy & Security

Xibe Chat is designed with privacy as a core principle. Your data stays on your device.

Local-First Architecture

  • All conversations stored locally in SQLite database
  • No cloud sync or external data collection
  • Complete control over your data - export or delete anytime

API Key Security

Your API keys are stored securely on your device:

  • • Never transmitted to Xibe Chat servers (we don't have any!)
  • • Used only to authenticate directly with AI providers
  • • Stored in encrypted local storage

Open Source Transparency

Xibe Chat is 100% open source. You can review the code on GitHub to verify our privacy claims. Read our full Privacy Policy.

Code Execution

Execute code directly in the chat using the E2B sandbox environment.

Supported Languages

🐍

Python

📜

JavaScript

💻

Bash

Custom API Providers

Add your own API providers to use custom AI models with your API keys. Xibe Chat supports OpenAI, Anthropic, OpenRouter, and any OpenAI-compatible API.

Built-in Providers

OpenAI

GPT-4, GPT-3.5, and other OpenAI models

https://api.openai.com/v1

Anthropic

Claude 3.5 Sonnet, Haiku, and more

https://api.anthropic.com/v1

OpenRouter

Access to 100+ models from various providers

https://openrouter.ai/api/v1

Groq

Ultra-fast inference for open models

https://api.groq.com/openai/v1

How to Add a Provider

  1. 1Navigate to Settings → Custom Providers → API Providers
  2. 2Select a built-in provider or click + to add custom one
  3. 3Enter your API key for the provider
  4. 4Click the list icon to add models
  5. 5Configure model details: name, ID, capabilities
  6. 6Use your custom model in any chat!

Example: Adding OpenAI GPT-4

1. Edit OpenAI provider and add your API key

API Key: sk-proj-xxxxxxxxxxxxx

2. Add a model:

Display Name: GPT-4 Turbo
Model ID: gpt-4-turbo-preview
Description: Most capable GPT-4 model
Supports Vision: ✓
Supports Streaming: ✓

Security Note

API keys are stored locally and encrypted. Never share your API keys with others.

MCP Servers

Model Context Protocol (MCP) servers extend AI capabilities with custom tools and data sources.

Configuration

Configure MCP servers in Settings → MCP Servers

Example configuration:

{
  "filesystem": {
    "command": "npx",
    "args": ["-y", "@modelcontextprotocol/server-filesystem", "/path/to/dir"]
  }
}

Building from Source

Prerequisites

  • • Flutter SDK (3.0+)
  • • Dart SDK
  • • Android SDK (for Android builds)
  • • Xcode (for macOS/iOS builds)

Build Commands

Android:

flutter build apk --release

Windows:

flutter build windows --release

Linux:

flutter build linux --release

macOS:

flutter build macos --release

Contributing

Contributions are welcome! Here's how to get started:

  1. 1Fork the repository on GitHub
  2. 2Create a feature branch: git checkout -b feature/amazing-feature
  3. 3Make your changes and commit: git commit -m 'Add amazing feature'
  4. 4Push to your fork and submit a Pull Request

Resources

Need More Help?

Can't find what you're looking for? Check out our GitHub repository or open an issue.

GitHub Download Now