Pi
MarkdownTerminal agent · OpenAI Chat, Anthropic Messages, Gemini
Full guide
| Format | Where to set it up |
|---|---|
| OpenAI Chat, Anthropic Messages, Gemini | ~/.pi/agent/models.json, or the project’s .pi/models.json after trusting the project; select a model with /model |
Setup
Copy the configuration and replace sk-... with your key from the Ailoot Dashboard. Prefer storing the key in the environment variable AILOOT_API_KEY, if the application supports it.
Before starting, request GET /v1/models with your Ailoot key. Choose an exact model ID whose methods include POST /v1/chat/completions, POST /v1/messages, POST /v1beta/models/{model}:generateContent, then use that ID in the configuration. For agents, also check tool support and the features required by your client version.
{
"providers": {
"ailoot": {
"baseUrl": "https://api.ailoot.ai/v1",
"api": "openai-completions",
"apiKey": "$AILOOT_API_KEY",
"models": [
{
"id": "claude-sonnet-5-5",
"input": [
"text",
"image"
]
},
{
"id": "deepseek-v4-pro"
}
]
}
}
}This configuration reads the key from the environment variable AILOOT_API_KEY. Set it before starting the application:
export AILOOT_API_KEY="sk-..." # add to ~/.zshrc or ~/.bashrcsetx AILOOT_API_KEY "sk-..." # then restart the terminalExample model: claude-sonnet-5-5. Choose any other model from the catalog — check that it supports the format you need.
Good to know
Test
- For the OpenAI format, the URL is
https://api.ailoot.ai/v1(or the full path if the application asks for a full URL). - For the Anthropic format, the URL is
https://api.ailoot.ai— the API origin. The example above does not show the full setup for this format; consult the client’s documentation for the path to use in the relevant field. The API route prefix is/v1. - For the Gemini format, the URL is
https://api.ailoot.ai— the API origin. The example above does not show the full setup for this format; consult the client’s documentation for the path to use in the relevant field. The API route prefix is/v1beta. - Ask a short question. If you get an error, check the code and Request ID, then consult Errors and limits.
- 404 immediately after setup almost always means the URL is wrong: an extra or missing
/v1. - 402 — insufficient balance for the reserve. Agents reserve more than chat: why.
Application documentation: https://github.com/earendil-works/pi/blob/main/packages/coding-agent/docs/models.md