Hollama: Browser-Based LLM Chat

Hollama is a minimal, privacy-focused Large Language Model (LLM) chat application that runs entirely in your browser. It's designed to connect to local or remote AI model servers like Ollama and OpenAI, providing a clean and powerful interface for interacting with your favorite models.

All your data—sessions, server configurations, and preferences—is stored locally in your browser, ensuring your conversations remain private.

Live Demo | Desktop Releases | Self-Hosting Guide

Hollama Chat Session

Key Features

  • Multi-Server Support: Connect to both Ollama and OpenAI (including compatible servers like llama.cpp) simultaneously.
  • Rich Model Support: Works with both text and vision models, allowing you to attach images to your prompts.
  • Full Markdown Rendering: Enjoy beautiful formatting for your conversations, including syntax highlighting for code and math notation rendering with KaTeX.
  • Advanced Chat Controls:
    • Code Editor: A dedicated editor for crafting complex prompts with code.
    • Fine-Tuning: Customize system prompts and adjust advanced Ollama parameters like temperature, Top-K, and more.
    • Message Management: Easily copy code snippets, full messages, or entire sessions. Edit and retry messages to refine your conversations.
  • Local First: All data is stored locally in your browser's Local Storage. You can easily import and export your data.
  • Model Management: Download Ollama models directly from the Hollama UI.
  • Customizable Interface: Features a responsive layout with both light and dark themes, and a multi-language interface.

Get Started

Ready to dive in? Head over to the Installation guide to learn how to use the live demo, install the desktop app, or self-host your own instance.