JARVIS RD Assistant

JARVIS RD Assistant is an open source, self-hosted workspace for researchers who want to search scientific literature and trace AI-generated answers back to papers. It runs on your own hardware with Ollama and connects literature review with PDF annotation, learning cards and research projects. It uses the Apache 2.0 license.

The Ask interface answers questions across your library with inline citations and exact source passages. Summaries and flashcard evidence also link to retrieved text, so you can inspect the basis for a claim. These checks establish traceability, not independent fact-checking; generated answers can still be wrong. Search combines full-text and vector retrieval to find relevant passages.

Paper discovery draws on arXiv, OpenAlex, Semantic Scholar and PubMed. Pulse presents ranked daily briefings and topic digests, while your saves and ratings influence later recommendations. Zotero sync connects the library to your citation manager. Tasks and milestones can link directly to source papers, keeping project work alongside the literature it depends on.

Learning cards turn paper findings into spaced-repetition reviews using FSRS scheduling, with retention analytics. An optional Telegram bot brings digests, paper triage, project tasks and card review into chat.

With local inference, your library and questions can stay on infrastructure you control. Optional cloud models connect through LiteLLM and receive the prompts and source excerpts needed for a task. Cached reading and review can work offline; discovery feeds need external APIs, and generation needs the configured model backend.