Continue
AI autocomplete and chat right inside VS Code – with local models.
Continue is an extension for VS Code and JetBrains. It completes code as you type and answers questions about your project. In Continue's settings, you pick Ollama as the provider and, for example, qwen2.5-coder as the model.
What is Continue?
Continue is an extension for the VS Code code editor. There's also a plugin for JetBrains and a command-line program (CLI). The extension helps you while you code. It completes code as you type. It also answers questions about your project in a chat right inside the editor.
The README describes Continue as an open-source coding agent. The source code is licensed under Apache-2.0, so you can use the tool for free.
How does it work with local models?
Continue doesn't ship with its own AI. It connects to a language model. That's the program that suggests the code or answers your questions. This model can run locally on your machine.
To do that, you choose Ollama as the provider in Continue's settings. Ollama is a program that runs language models on your own computer. For the model, you might pick qwen2.5-coder, which is built for programming.
Important for privacy: with a local model, your code stays on your machine. That applies to the model requests themselves. The README also states that the final version has removed anonymous telemetry. Telemetry means the program sends usage data to the vendor.
What does it do especially well?
- It makes suggestions right as you type.
- It answers questions about your project.
- You can freely choose and switch the model.
- It runs in VS Code and JetBrains, and there's also the CLI.
What hardware do you need?
For autocomplete, the small qwen2.5-coder:1.5b model is enough. According to the description, it runs almost anywhere. There's no information here about larger models for chat. As a rule of thumb, larger models need more memory and compute.
What are the limits?
The README says the continuedev/continue repository is no longer actively maintained and is read-only. There was a final version, 2.0.0, for VS Code, the CLI, and JetBrains. So don't expect new features or bug fixes. For JetBrains, the developers recommend using the CLI instead.
Small models give simpler suggestions than large ones. How good the results are depends on the model you choose.
What do you need?
- VS Code (or JetBrains)
- the Continue extension, available in the VS Code Marketplace or the OpenVSX Registry
- Ollama with a model like qwen2.5-coder
You'll find the documentation at docs.continue.dev.
Ideas: What you can do with it
You set up qwen2.5-coder:1.5b via Ollama as the autocomplete model. Then you write code in VS Code as usual and get suggestions. That saves typing on routine code.
You open the chat in Continue and ask what a particular part of your project does. That helps when you're reading unfamiliar or old code. You get an explanation without leaving the editor.
You're writing a script or a small app for fun and don't want to send any code to online services. With a local model via Ollama, everything stays on your machine.
If you're learning to program, you can ask in the chat why a line is written the way it is. The suggestions as you type also show you common ways of writing things. Always double-check the answers yourself, though.
No powerful graphics card? According to the description, the small qwen2.5-coder:1.5b model for autocomplete runs almost anywhere. That way, you get AI help even on weak hardware.
How to try it
Commands from the official guides – just copy them.
Linux
code --install-extension Continue.continue
ollama pull qwen2.5-coder:1.5b
ollama pull qwen2.5-coder:7bmacOS
code --install-extension Continue.continue
ollama pull qwen2.5-coder:1.5b
ollama pull qwen2.5-coder:7bWindows (PowerShell)
code --install-extension Continue.continue
ollama pull qwen2.5-coder:1.5b
ollama pull qwen2.5-coder:7bTips
- Start with qwen2.5-coder:1.5b for autocomplete. It's small and responds fast.
- Choose Ollama as the provider in Continue's settings. Check first that Ollama is running.
- Read the documentation at docs.continue.dev if you want to customize the settings.
- Check every suggestion before you accept it. AI-generated code can contain errors too.
- If you use JetBrains, try the CLI. The developers recommend it over the plugin.
If something goes wrong
- No suggestions appear: Check that Ollama is running and the model is loaded. In Continue, verify that Ollama is set as the provider and the model name is correct.
- Suggestions are slow: The model is probably too big for your machine. Switch to qwen2.5-coder:1.5b for autocomplete.
- You expect new features: According to the README, the project is no longer actively maintained. Version 2.0.0 is the final release.
- You can't find the extension in your editor: Search for "Continue" in the VS Code Marketplace or the OpenVSX Registry. For editors without the Microsoft marketplace, OpenVSX is the alternative.
Ask the AI assistant
Questions about Continue, installation or background? Our assistant knows this page and runs on our own machine, not in someone else's cloud.
The Assistant is an AI (Gemma 4) and can make mistakes – please check important details against the source. Questions are used only to generate an answer and are deleted after one hour. Please don't enter any personal data.
Comments
What do you think? Already tried it? Write it here, no sign-up needed.
Comments are public. Links aren't allowed, and we delete abusive posts. You can delete your own comment at any time in this browser. Learn more in the Privacy Policy.