hollama

Why we included this project

Hollama is a browser-based client for the Ollama and OpenAI APIs, so you can hold a conversation against models you already have installed without standing up another app. It stores sessions locally with import/export, which keeps your chat history on your own machine rather than a third-party service. The feature set covers what you actually want from a local-model front-end: multiple servers at once, vision and reasoning models, editable and retryable messages, markdown with syntax highlighting, KaTeX math, and custom system prompts. You can even download Ollama models straight from the interface instead of hunting them down on the command line. It self-hosts in a single Docker container, and there's a live demo that needs no sign-up if you want to try it before running anything.