pi (pi-coding-agent) | Northern Inference docs

Use Northern Inference with the pi coding agent over exact sovereign Canadian routes for government and regulated teams.

Northern Inference is a sovereign Canadian AI API for government, defence, health and regulated teams, including organizations under the Controlled Goods Program. Every route carries PIPEDA-aligned Canadian data residency, an exact path to the latest frontier and open-weight models, and a per-request chain-of-custody record.

Before you start

Create an API key. In the portal, open Keys and click Create Key. Your key starts with ni_live_.
Add credit. New accounts start with a balance of $0, and every request is refused until you add credit. Open Billing and click Buy credits.
Pick a model. Northern Inference model names end in a tier tag, for example claude-sonnet-4-6-t4. See them all on the Models page.

Install pi

Mac and Linux:

curl -fsSL https://pi.dev/install.sh | sh

Windows PowerShell:

powershell -c "irm https://pi.dev/install.ps1 | iex"

The installers also set up Node.js if you need it. If you prefer npm, run npm install -g --ignore-scripts @earendil-works/pi-coding-agent (Node.js 22.19 or newer). The older @mariozechner/pi-coding-agent package is deprecated.

Connect pi to Northern Inference

Create the file ~/.pi/agent/models.json (on Windows, C:\Users\you\.pi\agent\models.json) and paste this, with your key in place of ni_live_YOUR_KEY_HERE:

{
  "providers": {
    "northerninference": {
      "baseUrl": "https://northerninference.ca/v1",
      "api": "openai-completions",
      "apiKey": "ni_live_YOUR_KEY_HERE",
      "models": [
        { "id": "claude-sonnet-4-6-t4" },
        { "id": "DeepSeek-V4-Flash-t4" }
      ]
    }
  }
}

Start pi, type /model, choose a model under northerninference and press Ctrl+S to make it the default for new sessions. Send a message. It appears in the portal under Usage within a few seconds.

Each id is sent to Northern Inference exactly as written, so it must match a name on the Models page, including its case. To add a model, add another { "id": "..." } line.

Make sure you are using Northern Inference

In /model, always pick an entry under northerninference. If you have also signed in to another provider in pi, its models appear in the same list, and picking one of those sends your request to that provider instead.

If a model name is not recognized

Northern Inference answers with HTTP 404 and Model 'X' is not a route this key can call. It never swaps in a different model. Use a name from the Models page, or type ni-models as a message to list the models your key can call.

Troubleshooting

What you see	What to do
Your Northern Inference models are missing from `/model`	pi shows a model only when it can read its key. Check that `models.json` is valid JSON (quotes, commas, braces) and that the field is spelled `apiKey`.
401	The key is wrong or was revoked. Paste it again, without extra quotes or spaces.
402 `Insufficient credits. Add credits in the portal under Billing (https://northerninference.ca/portal/?page=billing) to continue.`	Add credit in the portal under Billing.
403 `privacy_tier_not_allowed`	Your key is limited to some tiers, and this model is in another one. Pick a model your key allows, or change the key's allowed tiers in the portal under Keys.
400 `ambiguous_model_route`	The name is the same model on more than one route your key can call. Use one of the route names the message lists.
404 `Model 'X' is not a route this key can call.`	Use a name from the Models page, or type `ni-models`.

Keep your key safe

If your key has been pasted into a chat, a screenshot or a shared screen, revoke it in the portal under Keys, create a new one and put it in models.json.