Pi

Point Pi at a Gateway key.

Run this

Writes the configuration below, backs up anything it touches, and picks up only the models your key actually serves. Run it again after connecting another plan and it adds what is new.

Terminal

ren inference setup pi

Check it worked

Terminal

pi --provider ren --model anthropic/claude-sonnet-5 -p 'hello'

A reply. Both pools live in this one provider, so `--model` also accepts the Codex ids.

Configure it by hand instead

The same result, written yourself. Replace <YOUR_KEY> with your key.

Set `sendSessionAffinityHeaders`
It is off by default on this transport, and without it every turn lands on a different seat and re-pays the cache.

Claude models

~/.pi/agent/models.json

Adds Ren as a provider alongside whatever Pi already has configured.

~/.pi/agent/models.json

{
  "providers": {
    "ren": {
      "baseUrl": "https://llm.useren.ai",
      "api": "anthropic-messages",
      "apiKey": "<YOUR_KEY>",
      "compat": {
        "sendSessionAffinityHeaders": true
      },
      "models": [
        {
          "id": "anthropic/claude-opus-5",
          "reasoning": true
        },
        {
          "id": "anthropic/claude-sonnet-5",
          "reasoning": true
        },
        {
          "id": "anthropic/claude-fable-5",
          "reasoning": true
        },
        {
          "id": "anthropic/claude-haiku-4-5",
          "reasoning": true
        }
      ]
    }
  }
}

ChatGPT models

~/.pi/agent/models.json

Adds Ren as a provider alongside whatever Pi already has configured.

~/.pi/agent/models.json

{
  "providers": {
    "ren": {
      "baseUrl": "https://llm.useren.ai",
      "api": "anthropic-messages",
      "apiKey": "<YOUR_KEY>",
      "compat": {
        "sendSessionAffinityHeaders": true
      },
      "models": [
        {
          "id": "openai/gpt-5.6-sol",
          "reasoning": true
        },
        {
          "id": "openai/gpt-5.6-terra",
          "reasoning": true
        },
        {
          "id": "openai/gpt-5.6-luna",
          "reasoning": true
        },
        {
          "id": "openai/gpt-5.5",
          "reasoning": true
        },
        {
          "id": "openai/gpt-5.4",
          "reasoning": true
        },
        {
          "id": "openai/gpt-5.4-mini",
          "reasoning": true
        },
        {
          "id": "openai/gpt-5.3-codex-spark",
          "reasoning": true
        }
      ]
    }
  }
}