Pi

Requesty routing for the Pi Coding Agent

Pi is an AI coding agent that runs locally on your machine.

Using the Requesty integration, you can:

• Access 300+ models from OpenAI, Anthropic, Google, Mistral, and many other providers through one API key.
• Get automatic prompt caching on Anthropic models, reducing cost significantly on multi-turn conversations.
• Track and manage your spend in a single location.
• Apply fallback policies, load balancing, and latency routing to keep your agent responsive.

Prerequisites

• Pi installed on your machine. See pi.dev for installation instructions.
• A Requesty API key from the API Keys page.

Configuration

1. Install Pi

   Follow the instructions at pi.dev to download and install Pi on your machine.

2. Configure the models file

   Create or edit the models configuration file at ~/.pi/agent/models.json and add Requesty as a provider:

   {
     "providers": {
       "requesty": {
         "name": "Requesty",
         "baseUrl": "https://router.requesty.ai",
         "api": "anthropic-messages",
         "apiKey": "rqsty-sk-....",
         "headers": {
           "HTTP-Referer": "https://pi.dev",
           "X-Title": "Pi"
         }
       }
     }
   }
   

   Replace rqsty-sk-.... with your Requesty API key from the API Keys page.

   With api: "anthropic-messages", the baseUrl must be https://router.requesty.ai without the /v1 suffix. Pi appends /v1/messages to the base URL for the Anthropic Messages format, so adding /v1 yourself points requests at /v1/v1/messages and they fail. Use https://router.requesty.ai/v1 only for the OpenAI Chat Completions format (api: "openai-completions").

   The HTTP-Referer: https://pi.dev header is what enables Requesty’s automatic prompt caching for Pi. When Requesty sees this referer on an Anthropic model, it automatically injects cache-control breakpoints so Pi’s large system prompt and tool definitions are reused across turns — cutting cost by up to 90% on long conversations. Combined with api: "anthropic-messages" (the native Anthropic Messages format), this gives you the best caching behavior. See Auto-Caching for details.

3. Install the Requesty extension

   Install the Requesty extension for Pi by running:

   pi install git:github.com/requestyai/pi-requesty
   
4. Sync models

   Run the following command inside Pi to sync the available Requesty models:

   /requesty-models-sync
   

   This pulls the latest model catalog from Requesty so you can select any approved model from your organization.

5. Reload the configuration

   Reload Pi to pick up the new provider and models:

   /reload
   

Selecting a model

After the sync completes, run the /model command inside Pi to choose a model from the list you synced.

You should be able to see both discrete models, e.g. anthropic/claude-opus-4-7 and your custom policies, e.g. policy/opus-europe.

Verifying the integration

Run a quick prompt inside Pi to confirm that requests are routed through Requesty. Then open the Requesty analytics dashboard to verify the request was logged. Spend, latency, and token counts should appear within seconds.

For caching, send a couple of follow-up messages in the same session on an Anthropic model, then check the request in the dashboard — after the first turn you should see cached tokens greater than zero, confirming the prompt prefix is being reused.

References

• Pi homepage
• Requesty Pi Extension
• Requesty API Keys
• Requesty Model Library
• Requesty Approved Models
• Requesty Auto-Caching