For developers

One API for every model

The idea: where the model runs doesn't change how you use it.

Point your code at localhost:7327 and pick a model. PearPie handles the routing. Your code stays the same whether the model is on your device, a peer's device, or on the PearPie Network.

Coming soon

The API is in the works

PearPie Chat is available now. The developer API is coming. Join the waitlist to be the first to know.

Join waitlist

What it will look like

curl http://localhost:7327/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{"model": "qwen-2.5", "messages": [{"role": "user", "content": "Hello!"}]}'

OpenAI compatible

If your code works with OpenAI's API, it works with PearPie. Change the base URL to http://localhost:7327 and you're done.

Works with any language

Python, JavaScript, Go, cURL, or any HTTP client. If it can make a POST request, it can talk to PearPie.