AI Chat Setup
This page documents ElastOS v1 (PC2), the proof of concept. ElastOS v2 is the current system. To install it, see Install ElastOS. To see how the stages fit together, see ElastOS in stages.
Configure AI assistants to work with your PC2 node.
Provider Options
Privacy: Maximum (all local) Cost: Free (your hardware) Speed: Depends on hardware
Ollama Setup (Recommended for Privacy)
Ollama runs AI models locally on your hardware.
Install Ollama
# Linux/macOS
curl -fsSL https://ollama.com/install.sh | sh
# Or download from https://ollama.comPull a Model
# Recommended: DeepSeek R1 (7B)
ollama pull deepseek-r1:7b
# Smaller/faster
ollama pull deepseek-r1:1.5b
# Larger/smarter
ollama pull deepseek-r1:14bModel Sizes
| Model | Size | RAM Required | Quality |
|---|---|---|---|
| deepseek-r1:1.5b | ~1GB | 4GB | Basic |
| deepseek-r1:7b | ~4.5GB | 8GB | Good |
| deepseek-r1:8b | ~5GB | 10GB | Better |
| deepseek-r1:14b | ~9GB | 16GB | Great |
| deepseek-r1:32b | ~20GB | 32GB | Best |
Configure PC2
- Go to Settings → AI Assistant
- Select Ollama as provider
- Ensure Ollama URL is
http://localhost:11434 - Select your model from dropdown
- Click Save
With Ollama, your conversations never leave your hardware.
Cloud Providers
Get API Keys
| Provider | Get Key At |
|---|---|
| OpenAI | platform.openai.com |
| Anthropic | console.anthropic.com |
| aistudio.google.com | |
| xAI | console.x.ai |
Add to PC2
- Go to Settings → AI Assistant
- Find your provider section
- Click Add Key
- Paste your API key
- Click Save
Available Models
OpenAI
- GPT-4o (recommended)
- GPT-4o Mini (faster/cheaper)
- GPT-4 Turbo
- GPT-3.5 Turbo
Anthropic (Claude)
- Claude Opus 4 (most capable)
- Claude Sonnet 4 (balanced)
- Claude 3.5 Haiku (fast)
Google (Gemini)
- Gemini 2.0 Flash
- Gemini 1.5 Pro
- Gemini 1.5 Flash
xAI (Grok)
- Grok 3
- Grok 3 Fast
- Grok 2
- Grok Vision
Using AI Chat
Sidebar
Click the AI icon in the taskbar or hover on the right edge.
Windowed App
Open “AI Chat” from the app launcher for full-screen experience.
Features
- Streaming — Watch responses generate in real-time
- Tool Calling — AI can read/write files on your node
- Multi-Model — Switch models mid-conversation
- History — Conversations persist across sessions
Tool Capabilities
The AI can:
- Read files — Analyze documents, code, data
- Write files — Create or modify files
- List directories — Browse your filesystem
- Execute commands — Run terminal commands (with permission)
⚠️
Tool calling requires explicit permission. The AI will ask before making changes.
Privacy Considerations
| Provider | Your Data |
|---|---|
| Ollama | Never leaves your hardware |
| Cloud Providers | Sent to provider’s servers |
For maximum privacy:
- Use Ollama
- Disable cloud providers in settings
- Your conversations stay 100% local
→ Next: Backup & Restore