.mobile-view{display:none}.docs-sidebar{display:block!important;position:static!important;width:auto!important;max-height:none!important}.docs-layout{grid-template-columns:1fr!important}You can read and navigate these guides without JavaScript. Enable JavaScript to use search, copy buttons and the playground.

VS CODE EXTENSION

Connect Roo Code to Outrun.

Connect Roo Code in VS Code to Outrun with your key and a model.

Roo Code was archived on May 15, 2026 and no longer gets updates. For a maintained version, use Zoo Code, a community fork; the same Outrun settings work there.

Before you start

Have your API key and some Outrun credit ready. Keep your key private.

Get your API key Check balance Choose a model

Optional setup helperUse an AI assistant to walk through these steps

Copy the setup instructions into an AI assistant you already use. Review any file changes it suggests before allowing them.

Enter your API key directly in the app or local settings. Keep it out of the assistant’s chat.

Connection settingsShow

Connection settings

API Provider: OpenAI Compatible
Base URL: https://api.outrun.to/v1
Model: claude-sonnet-5-5

Get your key from Outrun → API keys and enter it privately in the app or local settings. Test with “Reply with OK only.” Tests use credit; check Activity before retrying.

Setup Steps

1
Get your Outrun key
Open API keys on the Outrun website, or open @outrunAPI_Bot and choose My API key. Paste your key only into Roo Code's API Key field.
2
Install Roo Code
In VS Code, open Extensions, search for Roo Code and install it (extension ID RooVeterinaryInc.roo-cline). Then open the Roo Code panel from the activity bar.
3
Open the settings
In the Roo Code panel, click the settings gear icon. Under API Provider, choose OpenAI Compatible.
4
Fill in each field
Base URL: https://api.outrun.to/v1. API Key: paste your own Outrun key. Model: type claude-sonnet-5-5 for a first test, or any Roo-ready model ID from the list below. If Roo Code does not list it, choose the Use custom option that shows your typed ID. Do not add /chat/completions to the Base URL.
5
Match the model settings
Turn off Enable R1 model parameters. Keep Enable streaming and Include max output tokens on unless the table below says otherwise for your model. Set Max Output Tokens and Context Window Size from the table below, and do not use -1 for Max Output Tokens.
6
Save and send a test
Click Save, then Done. Start a new task and send: Reply with OK only. Roo Code adds its own instructions and tool list to every request, so even this test uses more credit than a short chat message. If it is interrupted or unclear, check Activity before trying again.

Roo-ready models to start with

Roo Code needs a model that supports tool calling. These routes are marked Roo-ready in Outrun's model setup data. Open a model page for its full Roo Code settings.

Model	Model ID	Details
Claude Sonnet 5.5	`claude-sonnet-5-5`	Model details
GPT-6.1 Sol	`gpt-6.1-sol`	Model details
DeepSeek V4 Pro	`deepseek-v4-pro`	Model details
DeepSeek V4.1 Flash	`deepseek-v4.1-flash`	Model details

Model settings to enter

Enter these after you choose the model. They come from the same setup data as the Roo Code panel on each model page. Context Window Size is the model's context limit; Max Output Tokens is a safe response length for Roo Code.

If a model page lists a special setting, such as Enable streaming off, follow it. Leave Enable Reasoning Effort off unless you need it, and clear any separate max tokens override in your Roo Code profile because it replaces the value above.

Model ID	Context Window Size	Max Output Tokens	Enable streaming	Image Support
claude-sonnet-5-5	1,000,000 tokens	8,192 tokens	On	On
gpt-6.1-sol	1,050,000 tokens	8,192 tokens	On	On
deepseek-v4-pro	1,000,000 tokens	8,192 tokens	On	Off
deepseek-v4.1-flash	1,000,000 tokens	8,192 tokens	On	Off

Optional: Roo Code cost display

Roo Code can show an estimated cost if you fill in Input Price and Output Price. Your Outrun credit is charged at the price on the model page, not by this display. Prices as of 2026-10-29, per 1M tokens, for the first pricing tier. Larger prompts can fall into a higher tier, so check the model page.

Model ID	Input Price	Output Price	First pricing tier
claude-sonnet-5-5	$0.49	$2.45	All prompt sizes
gpt-6.1-sol	$0.1176	$0.588	up to 272k tokens
deepseek-v4-pro	$1.35	$4.05	All prompt sizes
deepseek-v4.1-flash	$0.3	$1.2	All prompt sizes

Test your key first

This optional cURL request separates a key or balance problem from an app setting. Put your key in OUTRUN_API_KEY privately before running it. One model request uses credit; check Activity before retrying.

Chat Completions test

curl https://api.outrun.to/v1/chat/completions \
  -H "Authorization: Bearer $OUTRUN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"claude-sonnet-5-5","messages":[{"role":"user","content":"Reply with OK only."}],"max_tokens":16}'

More models and guides

Any text model that supports tool calling can work in Roo Code. For uncensored and abliterated models, follow the dedicated guide.

Browse model pages Uncensored (Roo Code) guide API formats and endpoints

Common errors

401: key missing or rejected

Copy your current key from API keys and paste it into Roo Code's API Key field with no spaces. Save, then try again.

402: not enough available credit

Check Balance and the key's spending cap. Another request may have temporarily reserved credit; read the message before adding funds.

404: model or URL not found

Copy the model ID exactly from the list above or Models & Pricing. Also check the Base URL: it must end at /v1 with no /chat/completions.

429: busy or rate limited

Wait for the active request to finish or the rate limit to reset, then try again. Do not resubmit the same request repeatedly.

Model choices and service notes

Use a currently available text model that supports tool calling from Models & Pricing; paused models cannot be used. Tests and tasks use prepaid Outrun credit. Check Activity before repeating an interrupted request.

Contact @outrunsupport

Frequently asked questions

What Base URL do I use for Roo Code?

Use https://api.outrun.to/v1 with the OpenAI Compatible provider. Do not add /chat/completions; Roo Code adds the request path itself.

Which Roo Code provider should I choose?

Choose OpenAI Compatible under API Provider. It is the option that accepts a custom Base URL, an API key and a model ID.

Does the model need tool calling?

Yes. Roo Code relies on native tool calling, so pick a Roo-ready model from the list above. A model without tool calling cannot be used with Roo Code.

Why does a short test use so much credit?

Roo Code sends its own instructions and tool definitions with every request, so even a short message carries a large prompt. Check Activity for the actual usage.

Why shouldn't Max Output Tokens be -1?

Roo Code can send -1 literally instead of letting the server choose. Use the number from the table for your model.

Does Outrun store my prompts?

Outrun does not store your prompts, images or outputs unless you choose to save them. Roo Code may keep its own task history on your computer.