Skip to main content
AI Choice EngineAI Choice Engine
Browse decision tools
← All setup guides

Continue · Setup guide

How to set up Continue

Continue was acquired by Cursor (Jun 2026). Final OSS 2.0 still installs in VS Code for local BYOK/Ollama use, but Hub/cloud are wound down and the repo is read-only — treat this as a migration/archive path, not a new team standard.

Continue

editor plugin · checked August 8, 2026

~10 minBeginner
Start installOpen ContinueCompare itBuyer guide

The “Open Continue” link goes to the vendor site. It is not an affiliate-tracked link on this page — no commission is claimed for that click. Affiliate disclosure.

Before you begin

  • VS Code already installed (JetBrains plugin is community-maintained; prefer Continue CLI cn if you need JetBrains).
  • An API key for at least one model provider (or a local model server such as Ollama).
  • A project folder you can open safely.
  • Expectation that Hub marketplace blocks / paid Models Add-On paths may be dead after the Jul 15, 2026 cloud wind-down.

Authentication

Provider API keys via Continue UI or ~/.continue/.env secrets

Account: Model provider account and API key (or local Ollama)

IDE extensions often cannot see shell export variables — put secrets in ~/.continue/.env and reference them from config.yaml. For local models, use OpenAI-compatible endpoints: Ollama at http://127.0.0.1:11434/v1 or LM Studio at http://127.0.0.1:1234/v1. Hub/Continue account auth docs may be stale after the final release.

Install or open it

1. Get the harness ready

These steps are written for beginners. On Mac, Terminal is the black/white text window where you type commands. On Windows, use PowerShell or Terminal. Use the OS tabs to hide steps that are not for your computer, then copy one command at a time.

  1. 01

    Read the status before you install

    continue.dev now shows the Cursor acquisition notice. The extension can still run locally, but there is no active commercial roadmap. For new rollouts prefer Cline, OpenCode, Kilo, or Cursor.

    Continue homepage
  2. 02

    Open VS Code

    Launch VS Code from Applications (Mac) or the Start menu (Windows).

  3. 03

    Open Extensions

    Mac

    Press ⌘ + Shift + X (or click the Extensions icon).

  4. 04

    Open Extensions

    Windows

    Press Ctrl + Shift + X (or click the Extensions icon).

  5. 05

    Install Continue

    Search for Continue and install the extension published by Continue (Continue.continue). Prefer config.yaml + ~/.continue/.env over Hub block imports. Drag the icon to the right sidebar if you prefer chat on the right.

    Continue install docs
  6. 06

    JetBrains path (optional, demoted)

    Settings → Plugins → Marketplace → Continue → Install → restart. Maintainers recommend the Continue CLI (cn) beside JetBrains rather than relying on the plugin alone.

    Continue install docs
  7. 07

    Optional: install the Continue CLI

    Docs lead with the install script. For the preferred JetBrains-adjacent path, install the Continue CLI and open a fresh terminal when it finishes. The command is cn, not continue. `npm install -g @continuedev/cli` remains an alternate.

    bash
    curl -fsSL https://raw.githubusercontent.com/continuedev/continue/main/extensions/cli/scripts/install.sh | bash

    You should see: The installer finishes without an error. Open a new terminal before the verify step so PATH updates apply.

    If this fails
    • The install command errors, or the next verify step says “command not found”. Re-run the official install command with network access, then open a brand-new terminal window before verifying.
    Continue CLI quickstart
  8. 08

    Verify the Continue CLI

    Confirm the CLI is on PATH before using it beside JetBrains or from a repository folder.

    bash
    cn --version

    You should see: A version string. Open a new terminal or check npm’s global bin directory if the command is missing.

    If this fails
    • Continue installed but never responds. Confirm a model is selected, the API key is in ~/.continue/.env (not only a shell export), and restart the IDE after changing secrets.
    Continue CLI quickstart

Repository or project

2. Start in the right place

Keep the first repository small, clean, and easy to roll back while you learn the harness’s permissions and context model.

  1. 01

    Open your folder

    File → Open Folder and pick a practice project.

  2. 02

    Open Continue and add a model

    Click the Continue icon. Configure a provider manually in config.yaml (not Hub slugs). For local models, run ollama serve (port 11434) or LM Studio (port 1234) and point config at the /v1-compatible base URL; match model tags exactly.

    Continue FAQ — secrets

First run

Chat and explain only

Use this task to test the workflow before handing over a larger repository or production responsibility.

  1. 01Highlight a small function and ask Continue to explain it in plain language.
  2. 02Ask for a suggested improvement without applying an edit.
  3. 03Only then try an inline edit on a comment or docstring and review the diff.

Why this is safe: Continue’s chat/explain loop teaches context selection before agentic edits change behaviour.

Stop and review if

  • Autocomplete or agent mode starts rewriting large files without a clear diff.
  • API keys are written into a committed config file.
  • Local Ollama models are referenced but ollama serve is not running.

The operating boundary

Privacy, data, and cost

The setup is incomplete until you know what the harness can see, what account pays for it, and who approves a risky action.

Privacy

Continue’s open-source extension does not train on your code. Prompts go to models you configure; anonymous telemetry excludes code/prompts and can be disabled.

Training on code: no

Depends entirely on the configured providers or local models.

Sensitive repos: Use local models for sensitive code when possible; otherwise scrub secrets and confirm provider terms.

Pricing

The Continue extension remains free OSS; you pay providers or run local models. Former Hub / Models Add-On commercial tiers are effectively defunct after acquisition.

Model: usage based

Watch out: Misconfigured autocomplete models can quietly generate large billable traffic.

Estimate a seat stack

When setup goes wrong

  • Continue installed but never responds.

    Confirm a model is selected, the API key is in ~/.continue/.env (not only a shell export), and restart the IDE after changing secrets.

  • Hub marketplace blocks or Continue account login fail.

    Expected after Hub wind-down — switch to manual provider models in config.yaml, or migrate to Cline / OpenCode / Kilo / Cursor.

  • You expected an autonomous agent like Cline.

    Continue is primarily chat/edit/autocomplete. For VS Code agent loops on local or BYOK models, prefer Cline or Kilo; keep Continue only if you want assistive chat on a frozen OSS release.

  • JetBrains plugin missing after install.

    Restart the IDE, enable the Continue tool window, or use the Continue CLI (cn) as the supported JetBrains-adjacent path.

  • Local Ollama models never appear or hang.

    Run ollama serve on port 11434 (or LM Studio on 1234 with /v1), pull the exact tag named in config.yaml, and keep secrets in ~/.continue/.env as KEY=value (Continue interpolates ${{ secrets.KEY }} — a shell export alone is not enough).

Next steps

After the first successful run

Keep the pilot small, then compare surfaces and budget before a team rollout.

Compare harnessesBuyer guideLocal LLM coding setupOpen-source agents hubAI stack cost calculatorGPU inference (rent)Local GPU sizingBuyer guidesLocal open-source AI guide

Setup details change quickly. The page was checked August 8, 2026; use the official source below before installing in a managed or production environment.

Continue install docsCompare all harnesses