VS Code + Continue with Local Ollama Models
Set up the Continue extension in VS Code to autocomplete, chat, and edit with local Ollama models so your code never leaves your machine.
Continue is an open-source AI code assistant for VS Code. When paired with a local Ollama model, you get autocomplete, chat, and inline edits without sending code to a cloud API. This recipe installs and configures Continue to talk to Ollama.
- VS Code installed.
- Ollama installed and running locally (see Run Ollama on macOS with Apple Silicon or Run Ollama on Ubuntu 24.04 with NVIDIA CUDA).
- A local model pulled, such as
qwen3.5:9borqwen2.5-coder:14b.
- Continue cannot connect: Verify Ollama is running:
curl http://localhost:11434. - Autocomplete is slow: Use a smaller model or reduce the context window.
- Model quality is low: Try a larger coder model or quantize to fit your hardware.
Developers who want IDE-native AI assistance with full privacy and no API costs.
Steps
Open VS Code, go to Extensions, and search for Continue. Install the official Continue extension.
Open Continue's settings and add an Ollama config:
{
"models": [
{
"title": "Ollama Qwen 9B",
"provider": "ollama",
"model": "qwen3.5:9b",
"apiBase": "http://localhost:11434"
}
]
}
In Continue settings, enable autocomplete and point it at a fast local model:
{
"tabAutocompleteModel": {
"title": "Ollama Fast Autocomplete",
"provider": "ollama",
"model": "qwen2.5-coder:1.5b",
"apiBase": "http://localhost:11434"
}
}
Open a file in VS Code. Use the Continue chat panel to ask a question about your code. Try tab completion. Both should use the local model.
Recipe verified 2026-07-01. Commands are tested but your environment may differ.
Browse related services