BOAR review
A beta desktop app for cited, offline document Q&A with local models.
Reviewed by iTechGuides Editors · Editorial team · Updated Nov 2026
BOAR desktop is a beta, open-source research assistant for Windows and Linux. It runs language models and a searchable knowledge library locally, answering questions from Wikipedia packs or imported PDF, Markdown, text, CSV and JSON files while displaying source passages. That makes it a fit for solo and small-team users who want desktop research without an account after the initial downloads. The free open-source model keeps the product focused on local operation rather than a hosted subscription service.
Its standout capability is the combination of cited retrieval and flexible local inputs. Wikipedia knowledge packs can contain up to 50,000 articles, while watched folders automatically index new documents. Profiles can restrict collections and response style, and offline voice transcription plus read-aloud answers support hands-free research. GGUF model selection and Hugging Face search give users control over model sourcing. For developer workflows, BOAR can expose an OpenAI-compatible localhost API, work with the OpenAI SDK, and share selected collections through MCP with Claude Desktop, Cursor and other MCP clients.
The breadth is useful, but the product remains a desktop beta. Users must download models and libraries before working offline, and support is provided through documentation and community channels. PIN-protected encrypted chat storage adds a clear local security measure, while Windows and Linux coverage suits desktop deployments. BOAR is a strong choice for individuals or small teams handling private documents, building local knowledge collections or connecting local Q&A to compatible tools. Organizations needing a mature release, broader deployment targets or a conventional paid support model should consider an alternative.
BOAR pros and cons
- Where it wins
- Cited offline Q&A across imports and Wikipedia knowledge packs
- Local API, MCP sharing, voice tools and watched-folder indexing
- PIN-protected encrypted chats with profiles for collection control
- Where it doesn't
- Beta desktop release may not suit production-critical workflows
- Initial model and library downloads are required before offline use
- Supports Windows and Linux desktop deployment
BOAR fact sheet, pricing and score →
Advertiser disclosure: iTechGuides is reader-supported. Vendors can pay for top positions in our rankings and for a place on other products' pages, and we may earn a commission when you click some links. Bought places stand at the number bought; this page and our disclosure say so. Paying vendors may get a fuller, more detailed review from our editors, and that review can change their score. How we rank.
Last updated · How we research and update