How to deploy an open-source AI chat app on Hostinger VPS

Hosting your own AI chat application gives you full control over your data, your models, and your costs without relying on third-party services. This article covers five AI chat applications available in the Hostinger Docker catalog to help you find the right fit for your needs. Each application can be deployed on your Hostinger VPS with a single click.

NOTES

• Some applications on this list require only an API key from an LLM provider such as OpenAI.
• PrivateGPT runs local LLMs directly on your VPS and requires significantly more RAM — a minimum of 8 GB is recommended, with 16 GB or more for comfortable performance.

Choose an AI chat application

Open WebUI

Open WebUI is a self-hosted AI chat interface that supports multiple LLM providers and includes retrieval-augmented generation (RAG) capabilities, allowing you to chat with your own documents alongside standard AI conversations. It includes model switching, conversation history, and multi-user management.

• Best for – Users who want a polished, full-featured ChatGPT alternative with multi-model support.
• 1-click deploy – Open WebUI

LibreChat

LibreChat is an AI chat interface that connects to multiple LLM providers — including OpenAI and Anthropic — through a single interface. It supports RAG, conversation branching, and a plugin system, making it a strong choice for users who work across several AI providers.

• Best for – Users who want to manage multiple AI providers through one unified interface.
• 1-click deploy – LibreChat

LobeChat

LobeChat is a modern open-source AI chat framework with support for multiple LLM providers, plugins, custom agents, and text-to-speech. It places a strong emphasis on interface design and extensibility.

• Best for – Users who prioritize a modern interface and want to extend functionality with plugins and agents.
• 1-click deploy – LobeChat

Chatpad AI

Chatpad AI is a privacy-focused web interface for ChatGPT that stores all conversation data locally on your VPS. It is a lightweight, minimal option with a straightforward setup — no accounts or external services beyond your LLM provider.

• Best for – Users who want a minimal, privacy-first interface with local data storage.
• 1-click deploy – Chatpad AI

PrivateGPT

PrivateGPT is a RAG engine built for chatting with private documents using local LLMs. Unlike the other applications on this list, PrivateGPT runs entirely on your VPS — no data is sent to any external service, including your LLM provider. This makes it the strongest option when document privacy is the primary requirement, but it also demands the most resources.

• Best for – Users who need to query sensitive documents without sending any data outside their VPS.
• 1-click deploy – PrivateGPT

Compare AI chat applications

A list appears showing the target audience and deployment complexity for each application:

Once deployed, each application is accessible through your VPS IP address and the port assigned during setup. For a full deployment walkthrough, see the resources below.

Your AI chat application is now deployed and ready for configuration.