A private AI research assistant: ask a question, it searches the web through its own SearXNG, reads the sources and writes a cited report, with no API key needed.
This product is available in the build below. Open it for the operating system, cloud and version detail, or read its deployment guide.
| Product | Cloud | Operating system | Version | |
|---|---|---|---|---|
| Local Deep Research | Azure | — | Standard | View · Guide |
Local Deep Research is an open source, MIT licensed research assistant. You ask a question in plain language; it breaks the question into searches, runs them, reads the pages it finds and writes a report with numbered citations and a list of sources. Runs are saved to a searchable history, reports export as Markdown or PDF, and a document library lets it research your own files as well as the web. Each account's data lives in its own database, encrypted with that account's password.
It suits analysts, researchers and teams that want deep web research inside their own cloud account, either fully local with an open model or with a hosted model and their own API key for stronger results.
The cloudimg image arrives wired end to end: Local Deep Research is connected to a private SearXNG metasearch instance and to Ollama on the same VM, with the openly licensed IBM Granite 4.0 Micro model (Apache 2.0) baked in, so research works the moment it boots with no key. Open self registration is switched off and first boot creates a single account with a password generated uniquely for each VM, the app is served over HTTPS by nginx and every internal service listens on loopback only. Every component is pinned and hash verified at build time, and each deployment is paired with a step by step deploy guide and backed by 24/7 support.