Deploy Local Deep Research

AI research assistant: cited reports from the web, arXiv and PubMed

Deploy Local Deep Research

Deploy and Host Local Deep Research on Railway

Local Deep Research is an open-source AI research assistant. You ask a question, and it plans searches, reads the sources it finds on the web, arXiv, PubMed, Wikipedia and your own documents, and writes a cited report. It works with cloud LLMs (OpenAI, Anthropic, Google, OpenRouter) and with self-hosted models served by Ollama or llama.cpp.

About Hosting Local Deep Research

This template deploys Local Deep Research 2.0.1 with a private SearXNG metasearch instance, so web search works without any search API key. The web app listens on Railway's PORT behind HTTPS and keeps everything (the user list, each user's SQLCipher-encrypted database, research reports, caches and logs) on a volume at /data. On first boot it creates one account from ADMIN_USERNAME and a generated ADMIN_PASSWORD, and open sign-up stays off. SearXNG has no public domain and is reached only over Railway's private network. After logging in, choose your LLM provider and paste its API key under Settings.

Common Use Cases

  • Literature reviews and background research with cited sources from arXiv, PubMed and the web
  • A private research assistant that searches your own uploaded documents alongside the web
  • Scheduled news briefings and follow-up research on topics you track

Dependencies for Local Deep Research Hosting

  • An LLM: an API key for a cloud provider (OpenAI, Anthropic, Google, OpenRouter and others) or the URL of your own Ollama or OpenAI-compatible server
  • SearXNG (included in this template, private network only)

Deployment Dependencies

Implementation Details

  • Log in with ADMIN_USERNAME (default admin) and the ADMIN_PASSWORD shown in the service's Variables tab. The password is also the key of your encrypted database and can't be recovered, so keep it safe. Change it in the app, not in the variable: the variables are used only once, to create the first account.
  • No LLM is preset, because a variable would lock the setting in the UI. Open Settings, pick a provider (for example OpenAI or OpenRouter) and paste its API key, or point it at your own Ollama server.
  • To let other people create accounts, set LDR_APP_ALLOW_REGISTRATIONS=true on the app service.
  • The SearXNG search URL is fixed to the private service (LDR_SEARCH_ENGINE_WEB_SEARXNG_DEFAULT_PARAMS_INSTANCE_URL). Other search engines (arXiv, PubMed, Wikipedia, Brave, Tavily and more) can be turned on in Settings.
  • Ollama isn't included: CPU-only Railway services are too slow for local models. Use a cloud provider or a GPU host of your own.

Why Deploy Local Deep Research on Railway?

Railway is a singular platform to deploy your infrastructure stack. Railway will host your infrastructure so you don't have to deal with configuration, while allowing you to vertically and horizontally scale it.

By deploying Local Deep Research on Railway, you are one step closer to supporting a complete full-stack application with minimal burden. Host your servers, databases, AI agents, and more on Railway.


Template Content

More templates in this category

View Template
Chat Chat
Chat Chat, your own unified chat and search to AI platform.

okisdev
117
View Template
Weaviate
Vector database with hybrid search and an API key-protected API

GitHub Locker
1
View Template
Möbius Personal AI OS
Self-hosted AI agent that builds and runs apps for you, with memory.

bento
1