LocalGPT is a private, on-premises document intelligence platform for questioning, summarizing, and analyzing files with locally hosted language models. Its data remains on the user’s machine, making it suitable for confidential or offline workflows. The retrieval system combines semantic similarity, keyword matching, late chunking, contextual enrichment, and sentence-level pruning. A smart router chooses between retrieval-augmented generation and direct model responses for each query. An independent verification pass is intended to improve answer reliability before results are returned. The platform works with Ollama models, offers a browser interface and API, and can run through local or Docker-based setups. It supports CPU and several hardware acceleration environments while retaining conversation history within a session.
Features
- Private on-device document processing
- Hybrid semantic and keyword retrieval
- Automatic RAG or direct-answer routing
- Context enrichment and pruning
- Local browser interface and API
- CPU, GPU, HPU, and MPS support