Ollama Client

Ollama Client is an open-source browser extension designed to enable users to interact with local large language models (LLMs) directly within Chrome and Firefox. cpp, as well as user-added OpenAI-compatible endpoints and Anthropic, allowing flexible integration with various AI providers. The extension prioritizes local-first operation, ensuring that no mandatory cloud API is required and giving users control over their data and endpoints.

Among its features, Ollama Client facilitates multi-chat sessions with branch navigation and local persistence using SQLite. Users can upload files in PDF, TXT, MD, and DOCX formats for retrieval-augmented generation (RAG), with configurable file size limits and background embedding progress. The extension includes enhanced content extraction capabilities through Defuddle integration, offering smart scroll strategies and specialized helpers for GitHub and YouTube. It also supports prompt templates for tasks like summarization, translation, and code assistance, with the option to create custom templates.

The tool provides optional web search functionality for models capable of tool use, integrating with search engines such as SearXNG, Brave Search, or Tavily, and allowing backend selection in the settings. Text-to-speech features include searchable voices, adjustable pitch and rate, and per-message playback with cross-browser voice loading. Users can back up and restore their data via ZIP snapshots with versioned manifests, and privacy controls are visible within the extension, enabling review, backup, wiping, chat tagging, and message branch forking without leaving the interface.

Advanced configuration options let users control provider base URLs, map selected models, check capabilities, and enable debug logging. js, and Defuddle. Its design emphasizes privacy and user control, making it suitable for individuals seeking to run and manage AI chat sessions locally within their browser environment.