A deep research skill and pipeline for Claude Code and OpenAI Codex that gathers web and scholarly sources into a persistent, searchable markdown vault and writes cited reports.
AnotiaWang/deep-research-web-ui
Self-hosted AI deep-research web app that plans searches, reads real sources and returns a report with citations you can inspect and follow up on.
Open-source alternative to:Perplexity
ChatGPT Deep Research
- Stars
- 2,225
- Forks
- 306
- Open issues
- 17
- Last commit
- 12 days ago
What is it?
Deep Research Web UI takes a research question and produces a cited report. It plans search queries, browses the web, checks findings against source pages, and lets you open the evidence behind each citation in the report.
It runs in the browser (Client Mode, where users enter their own API keys) or on a server (Server Mode, where keys are set as environment variables and users don't need them). Server Mode runs as a Docker image or a Nuxt SSR deployment. Supported AI providers include OpenAI-compatible endpoints, DeepSeek, OpenRouter, SiliconFlow, Ollama and LiteLLM. Supported search providers include Tavily, Firecrawl, Google PSE, You.com, Serply and fastCRW.
Findings can be followed up: a new search updates only the report blocks that cite a given finding, and the result is saved as a separate history entry so the original report is kept. Reports can be exported as Markdown or PDF.
Key features
- 01Plans search queries and streams progress to the UI in real time
- 02Shows the research process as a search tree
- 03Lets you inspect source excerpts behind each citation
- 04Follow-up research updates only affected passages and keeps the original report in history
- 05Supports Client Mode with keys in the browser and Server Mode with server-side keys
- 06Exports reports as Markdown or PDF
Use cases
- Producing a sourced overview of an unfamiliar topic with links to the pages used
- Running research on a self-hosted server so API keys stay off user devices
- Verifying specific claims in a report by following up on their citations
- Using local models through Ollama or LiteLLM with a web search provider
Quick start
- 1
Run the pre-built Docker image
docker run -p 3000:3000 --name deep-research-web -d anotia/deep-research-web:latest - 2
Open http://localhost:3000 in your browser and enter your own API keys (Client Mode)
Things to know
- The project has no license, so reuse and redistribution terms are unclear.
- Client Mode requires you to supply your own AI and web search API keys.
- Server Mode needs an SSR/Nitro runtime and cannot run as a purely static deployment.
In collections
Topics
FAQ
- What is deep-research-web-ui?
- Self-hosted AI deep-research web app that plans searches, reads real sources and returns a report with citations you can inspect and follow up on. Deep Research Web UI takes a research question and produces a cited report. It plans search queries, browses the web, checks findings against source pages, and lets you open the evidence behind each citation in the report.
- Is deep-research-web-ui free to use? What license does it have?
- deep-research-web-ui has no declared license. Its code is public, but without a license you have no explicit permission to reuse it; ask the author before relying on it.
- Is deep-research-web-ui still maintained?
- Actively maintained. The last commit to deep-research-web-ui was on Oct 26, 2026.
- What can I use deep-research-web-ui for?
- Producing a sourced overview of an unfamiliar topic with links to the pages used Running research on a self-hosted server so API keys stay off user devices Verifying specific claims in a report by following up on their citations Using local models through Ollama or LiteLLM with a web search provider
- How do I install deep-research-web-ui?
- Run the pre-built Docker image: docker run -p 3000:3000 --name deep-research-web -d anotia/deep-research-web:latest Open http://localhost:3000 in your browser and enter your own API keys (Client Mode)
- Where can I get deep-research-web-ui?
- The source code and installation instructions are on GitHub: https://github.com/AnotiaWang/deep-research-web-ui
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
AI-powered conversational search engine that answers questions with live web results, supporting OpenAI, Anthropic, Gemini and Vertex AI models.
An open-source autonomous research agent that plans questions, gathers web and local sources, and writes cited reports with any LLM provider.
An open-source, agentic AI research search engine that plans sub-tasks, retrieves live sources, and returns answers with inline citations.
A self-hosted metasearch engine with an AI assistant that can run entirely inside the browser, with no API key required.
Self-hosted AI research assistant that runs deep, multi-step searches across local and cloud LLMs and many search engines, with citations and encrypted storage.