Tessera
A self-hosted AI literature review workspace that stores papers locally, uses OpenAI-compatible backends and exports research websites.
Tessera is a self-hosted literature review workspace for researchers who want to find papers, keep reading notes and compare findings in one browser interface.
- Paper search: Search Semantic Scholar, arXiv, OpenAlex, CrossRef and PubMed together, with automatic duplicate removal and full-text search across saved abstracts and extracted PDF text.
- Collections and PDF notes: Group papers into collections, keep per-paper notes and annotate uploaded PDFs with highlights and comments.
- AI summaries and synthesis: Generate paper summaries covering findings, methods and limitations, then compare a collection's papers through thematic analysis and research gaps.
- Knowledge and citations: Browse extracted research details linked to their source papers, add your own conclusions and explore citation graphs with AI classifications of supporting, contradicting or mentioning relationships.
- Research export: Export citations as BibTeX, save papers, syntheses and knowledge in JSON bundles, or create a self-contained static website with summaries, synthesis and an interactive citation graph.
The app stores its research library in local SQLite storage and keeps papers and PDFs locally. AI analysis requires an OpenAI-compatible endpoint, which can run locally through Ollama, vLLM or NVIDIA NIM. Choosing a hosted provider such as OpenAI sends analysis requests to that service. Paper discovery needs internet access but doesn't require search API keys. Exported websites open locally or run on a static web host.