QAnything
A self-hosted document Q&A system that supports offline use, Chinese and English questions, and default CPU operation on Windows, macOS and Linux.
Alternative to:
QAnything is a local knowledge base for people and teams who want to ask questions about their own documents. It supports offline operation and questions across Chinese and English document collections.
- Document Q&A works with PDF, Word, PowerPoint, Excel, email, Markdown, CSV, text files and images, as well as HTML web links.
- Cross-language answers let you ask in Chinese or English regardless of the language used in the files.
- Passage search uses BCEmbedding to retrieve relevant material and rerank results before producing an answer.
- Chat and retrieval modes support conversation without documents, retrieval-only searches and custom bots.
- API access lets other applications connect to the document question-answering system.
QAnything runs through Docker on Windows, macOS and Linux. Its default environment uses CPU processing without requiring a GPU, and it supports offline installation and use.
The project is open source under AGPL-3.0. The open-source edition uses QwenLM, whose separate model license also applies to commercial use.
