Pi | Yolo Router Docs

iBefore you start

Pi is installed and can start normally.
A working Yolo Router endpoint, for example https://api.yolorouter.com.
A Yolo Router API Key generated from the console.
The model ID you want to use, for example deepseek-v4-pro. It must exactly match the model ID in the Yolo Router console.

Pi is a minimal, extensible terminal coding framework. It supports custom OpenAI-compatible providers through its config file. For Yolo Router, set baseUrl to the /v1 endpoint and use openai-completions.

Setup

Check Pi

Confirm that Pi is installed and can start:

npm install -g @mariozechner/pi-coding-agent

Set the API Key env var

The examples below use YOLOROUTER_API_KEY; the Pi config will read this variable.

export YOLOROUTER_API_KEY="sk-yolo-..."

$env:YOLOROUTER_API_KEY = "sk-yolo-..."

set YOLOROUTER_API_KEY=sk-yolo-...

Do not write the real API Key into config files or commit it to your repository.

Edit the Pi config file

Add a yolorouter provider to Pi's config file:

{
  "providers": {
    "yolo": {
      "baseUrl": "https://api.yolorouter.com/v1",
      "api": "openai-completions",
      "apiKey": "${YOLOROUTER_API_KEY}",
      "models": [
        {
          "id": "deepseek-v4-pro",
          "name": "deepseek-v4-pro",
          "contextWindow": 200000,
          "maxTokens": 128000,
          "input": ["text"],
          "reasoning": true,
          "compat": {
            "supportsDeveloperRole": false,
            "requiresReasoningContentOnAssistantMessages": true,
            "thinkingFormat": "anthropic",
            "reasoningEffortMap": {
              "minimal": "low",
              "low": "low",
              "medium": "medium",
              "high": "high",
              "xhigh": "max"
            }
          }
        }
      ]
    }
  }
}

Field notes:

baseUrl: your Yolo Router OpenAI-compatible endpoint, with /v1
api: always use openai-completions
apiKey: uses the $YOLOROUTER_API_KEY variable reference so you do not commit the real Key
models[].id: must exactly match the model ID in the Yolo Router console
compat.requiresReasoningContentOnAssistantMessages: keeps reasoning content on historical messages when thinking mode is active
reasoningEffortMap: maps Pi effort levels to the target model's supported reasoning levels; omit compat for non-reasoning models

Test the connection

Send a test message:

Introduce the model you are using in one sentence.

If Pi returns a normal response, the integration is working. You can also check the Yolo Router console for the corresponding request log.

Troubleshooting

yolorouter is not in the provider list

Check the Pi config file path and JSON format. Restart Pi after changing the config.

401 Unauthorized

Make sure YOLOROUTER_API_KEY is exported in the current shell and has no extra spaces.

Reasoning model acts strangely

Verify requiresReasoningContentOnAssistantMessages and thinkingFormat for reasoning models. For non-reasoning models, remove the compat block.

Model not found

Confirm that models[].id exactly matches the model ID in the Yolo Router console.

Pi

On this page