Does Pi work with Standard Compute?

Yes. Pi supports a custom OpenAI-compatible base URL, so it works with Standard Compute. Set the base URL to https://api.stdcmpt.com/v1, paste your API key, and set the model to "standardcompute".

What base URL do I use for Pi?

Use https://api.stdcmpt.com/v1 as the OpenAI-compatible base URL in Pi, with your Standard Compute API key and the model "standardcompute".

Do I need to change my code or model names?

No. The request and response format is identical to OpenAI's, and "standardcompute" is a single model identifier that routes to the best frontier model for each request — so you never pick or update model names.

← All integrations

OpenAI-Compatible · Custom Base URL

Terminal coding agent

Use Standard Compute with Pi

Mario Zechner's radically minimal open-source terminal coding agent — four core tools, a fast TUI, and TypeScript extensions for the rest. It speaks the OpenAI format, so it works with Standard Compute as a custom model provider — point the base URL at us, drop in your key, set the model to standardcompute, and get unlimited frontier-model compute at one flat price.

Power Pi with Standard Compute →

Your three settings

These are all Pi needs. Same values for every OpenAI-compatible tool.

Base URL

https://api.stdcmpt.com/v1

API Key

Your Standard Compute key

Model

standardcompute

Where to set them in Pi

The exact spot Pi keeps its OpenAI-compatible base URL.

Open ~/.pi/agent/models.json (create it if it doesn't exist).
Add a provider using the OpenAI API format with the Base URL below and your API key.
Add standardcompute as its model and select it in pi.

Pi docs ↗

Why point Pi at Standard Compute

The base-URL swap is the easy part — here's what you get on the other side.

Unlimited compute, flat price

No per-token billing or usage caps. One predictable monthly price, however hard your agent runs.

Frontier models, auto-routed

standardcompute routes each request to the best frontier model — you never pick or update model names.

Nothing to rewrite

Identical OpenAI request/response format. Swap the base URL and key; your code stays exactly the same.