What is the Pezzo Proxy?

The Pezzo Proxy is a proxy server that can be used to automatically proxy requests to LLMs through Pezzo.

Contrary to other solutions that force you to install SDKs and change the way you write code, we chose to build a proxy that can be used with any existing codebase. Here is why:

  • Easy to integrate: It allows you to integrate Pezzo into your codebase with zero changes to your LLM consumption code.
  • Always up-to-date: The proxy will never prevent you from using new features released by OpenAI. As soon as OpenAI releases new feautres, they are immediately supported by Pezzo.
  • Language agnostic: Whether you use Node.js, Python, Golang or any other language - Pezzo Proxy supports them all!
  • Enhanced LLM capabilities: The proxy layer allows us to build powerful features on top of LLMs, for example a caching layer that allows you to save money on your API calls.
  • Maintainable: We are able to push fixes to the Pezzo Proxy without requiring you to update any integration in your code base. It’s completely transparent.

Is the Pezzo Proxy open source?

Yes! As with any component at Pezzo, the Pezzo Proxy is open source under Apache 2.0 license.

For more information, you can refer to:

Next Steps