
Deploy DeepWiki-Open
AI wiki generator for GitHub/GitLab/Bitbucket repos, behind a login
Just deployed
/root/.adalflow
Deploy and Host DeepWiki-Open on Railway
DeepWiki-Open is an open-source (MIT) tool that turns a GitHub, GitLab or Bitbucket repository into an interactive wiki. It clones the code, builds embeddings, and uses an LLM such as OpenAI or Google Gemini to write documentation pages and architecture diagrams. You can then ask questions about the codebase in a chat.
About Hosting DeepWiki-Open
DeepWiki-Open runs a Python API (FastAPI) and a Next.js web UI. It also needs an API key for an LLM provider and disk space for cloned repositories, embeddings and generated wikis. The upstream app has no login, so this template puts a Caddy proxy in front of it. Caddy asks for a username and a generated password on every page except the health check. It also routes the UI, the API and the chat WebSockets through one HTTPS domain. A volume on /root/.adalflow keeps your wikis and indexes across redeploys. The service builds from a GitHub repo that pins the official image by digest, so Railway can notify you when the template is updated.
Common Use Cases
- Generate browsable documentation and diagrams for an unfamiliar or poorly documented codebase
- Ask questions about a repository in a chat ("Ask" and "DeepResearch") while onboarding new developers
- Keep a private, self-hosted alternative to hosted code-wiki services for internal or private repositories
Dependencies for DeepWiki-Open Hosting
- An OpenAI API key (or a Google Gemini key) for embeddings and wiki generation
- A volume for cloned repositories, embeddings and generated wikis (included)
Deployment Dependencies
- DeepWiki-Open source: https://github.com/AsyncFuncAI/deepwiki-open
- Official image: https://github.com/AsyncFuncAI/deepwiki-open/pkgs/container/deepwiki-open
- Template repo: https://github.com/aalfath/deepwiki-open-railway-template
Implementation Details
After the deploy finishes, open the public URL. Log in as admin (or the BASIC_AUTH_USERNAME you chose). The password is in the service's BASIC_AUTH_PASSWORD variable. With only an OpenAI key, the template makes OpenAI the default model. If you set GOOGLE_API_KEY, Gemini becomes the default. Other upstream settings, such as OPENROUTER_API_KEY, OLLAMA_HOST or DEEPWIKI_EMBEDDER_TYPE, can be added as service variables. If you change the embedder, existing repositories need to be re-indexed. To read private repositories, paste an access token in the UI when you generate a wiki.
Why Deploy DeepWiki-Open 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 DeepWiki-Open 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
OPENAI_API_KEY
OpenAI API key used for embeddings and wiki generation

