Local AI Workspace od LocalAI Workspace

Local AI Workspace is a powerful, privacy-first user interface designed for local large language models (LLMs) and retrieval-augmented generation (RAG). By connecting seamlessly to your locally running Ollama, LM Studio, or Llama.app instances, as well as custom OpenAI-compatible server APIs, it brings advanced AI capabilities directly into your browser sandbox with zero cloud data transmission.

Key Features:

100% Local & Private: Your notes, documents, and chat history never leave your machine. No external APIs, no tracking, and complete offline capability.
Multi-Engine Connectivity: Out-of-the-box support for Ollama, LM Studio, Llama.app, and any custom OpenAI-compatible server endpoints (like llama.cpp or vLLM).
Local-First Knowledge Base (RAG): Effortlessly drop your local Markdown notes (including Obsidian vaults), PDFs, and text files to build a highly optimized, client-side vector database via IndexedDB.
Hardware-Optimized Guidance: Features an integrated system architecture check that reads your local GPU configuration transparently via browser-safe WebGL APIs, offering tailored model recommendations for peak local performance.
Streamlined Interface: Experience seamless multi-mode switching between a General Assistant chat and rich Document-based Search without the complexity of command-line terminals.