Ollama Client
A privacy-first browser extension that adds a sidepanel chat workspace for local and remote language models, including Ollama, LM Studio and llama.cpp, with local retrieval-augmented generation and browser-context tools.
Ollama Client provides a browser-native sidepanel for chatting with local models through Ollama, LM Studio or a llama.cpp server, or with other OpenAI-compatible or Anthropic endpoints added from settings, with streaming responses, cancellation, image input for vision-capable models, and a selection-button overlay that captures highlighted page text into the chat. Tool-capable models can read the current tab, other open tabs, uploaded files and local memory, and optionally search the live web when a prompt calls for it, and uploaded files can be indexed for local retrieval-augmented generation using a shared or provider-native embedding model with hybrid keyword-and-dense search.
Chat history, sessions, files, settings and embeddings are kept on the user's machine by default, model routing falls back to Ollama when no saved mapping exists, and a dedicated privacy screen lets a user review all stored data, create a full backup, or wipe it.
It is written in TypeScript and released under the MIT license.