Quick Tip: Run Local LLMs with Ollama + Continue.dev in VS Code (Free Copilot Alternative)

You don't need a $10/month Copilot subscription to get AI code completion. Ollama + Continue.dev gives you local, private, free AI coding assistance in 5 minutes.

The Setup

# 1. Install Ollama
curl -fsSL https://ollama.com/install.sh | sh

# 2. Pull a coding model
ollama pull codellama:7b-code

# 3. Install Continue.dev extension in VS Code
# Search "Continue" in marketplace, install, then:

# 4. Configure Continue to use Ollama
cat > ~/.continue/config.json << 'EOF'
{
  "models": [
    {
      "title": "CodeLlama 7B",
      "provider": "ollama",
      "model": "codellama:7b-code"
    }
  ]
}
EOF

Why This Works

Feature	Copilot	Continue + Ollama
Price	$10/mo	$0
Privacy	❌ Cloud	✅ Local
Model choice	Locked	Any Ollama model
Offline	❌	✅
Context window	Limited	Full file + repo

Pro Tips

Autocomplete: Enable in Continue settings → Tab autocomplete → Ollama
Chat: Cmd+L (Mac) / Ctrl+L (Win) to open sidebar chat with your code context
Edit mode: Select code → Cmd+I → describe the change
Best models for coding: codellama:7b-code (fast), deepseek-coder:6.7b (better quality), starcoder:3b (lightweight)

The Catch

7B models are 80% as good as Copilot for boilerplate, but struggle with complex refactoring. For heavy lifting, switch to a 13B+ model or use the chat interface with more context.

I use MonkeyCode to configure my local AI stack: https://ly.cyberserval.tech/iIETXiF

What's your local coding model of choice?

ai #python #tips #opensource

DEV Community