Fine-Tune and Deploy LLMs On Your Own Infrastructure

💻 Quickstart   •   📄 Docs   •   💬 Discord

Haven lets you build LLM-powered applications hosted entirely on your own infrastructure.
Just select a model to run - Haven will set up a production-ready API server in your private cloud.

Getting Started 🔥

Setting up an LLM server requires just three steps:

1. Get an API key for a Google Cloud service account
2. Deploy Haven's manager container on a Google Cloud instance
3. Spin up a model worker using the Python SDK

A description of these steps can be found in our documentation.

Roadmap 🚀

We're constantly building new features and would love your feedback! Here's what we are currently looking to integrate into our platform:

• Inference Workers
• Google Cloud Support
• Fine-Tuning Workers
• AWS Support

Learn More 🔍

To learn more about our platform, you should refer to our documentation.