MemFree
Open-source AI search combines your knowledge base with web results. Self-host it with external APIs, or use the hosted service. Licensed under MIT.
Alternative to:

MemFree is an open-source AI search tool that answers questions using both your own knowledge base and the internet. It's for people who want to search saved documents, bookmarks and web content together, with AI summaries instead of reading each result separately. You can self-host it or use the hosted website.
Self-hosting still depends on external services. The documented setup uses OpenAI and Serper APIs plus Upstash Redis, so it isn't a fully offline system with local model inference. The project uses the MIT license.
Search accepts text, images, files and web pages. It supports PDF, Docx, PPTX and Markdown documents, and can compare or summarize multiple images. Chrome bookmark indexing brings saved links into your knowledge base. Contextual follow-up searches let you continue a question, and the tool can decide when it needs web results. Model integrations include ChatGPT, Claude and Gemini; search integrations include Google and Exa.
MemFree also includes an AI page generator for people building web interfaces. It creates React pages with Tailwind CSS and shadcn/ui from text, reference images or file content. You can preview the result, edit the generated code, and check responsive layouts and dark mode. The generator also supports screenshot export and publishing pages to the web.
Search results can include mind maps, images and videos alongside text. MemFree supports sharing results and syncing search history across devices.