PandaWiki
Self-hosted AI wiki from Chaitin, open source under AGPL
Overview
PandaWiki is a self-hosted knowledge-base system from Chinese security firm Chaitin (长亭科技). It pairs a Markdown and HTML editor with AI Q&A, AI search and embeddable chat widgets. The open-source edition is AGPL-3.0 and capped at one site and 300 documents; paid Professional, Commercial and Enterprise editions lift the limits.
Pricing
Pricing shown for reference only. These figures reflect RECATOOLS research as of 12 Oct 2026 and may be out of date or incomplete. This is not financial or purchasing advice — always confirm the current price on the provider’s official website before making any decision.
- 1 wiki site
- 300 documents per wiki
- 1 admin
- WeCom community support
- 10 sites, 10,000 documents per wiki
- 20 admins
- Multi-language sites
- Custom AI prompts and advanced SEO
- 50 admins
- SSO for up to 2,000 users
- MCP Server and API tokens
- Watermark and version history
- Commercial edition with limits removed
What you can produce with PandaWiki
- Self-hosting on Linux with Docker, or the hosted version on Baizhi Cloud
- AI search and Q&A grounded in your own documents
- Markdown and HTML editing with export to Word, PDF or Markdown
- Import from URLs, sitemaps, RSS feeds and offline files
- An embeddable chat widget plus DingTalk, Feishu, WeCom and Discord bots
- Paid editions that lift the free one-site, 300-document limit
PandaWiki suits teams that want AI-searchable docs or an FAQ without a SaaS subscription: self-host it with Docker, bring your own chat model and keep your data. Releases are frequent, and the backing company is an established, well-funded security vendor.
Check two limits first. The free edition allows one site, 300 documents and one admin, with SSO, version history, an MCP server and API tokens only in the paid Commercial edition. And the docs and community are almost entirely Chinese.
What we found
PandaWiki is a self-hosted knowledge-base system from Chaitin Tech (长亭科技), the Beijing security company behind the SafeLine web application firewall. It is meant for product documentation, help centres and internal FAQs with AI answers on top, and it runs on Linux under Docker. The code is on GitHub under AGPL-3.0, with a Go backend and a TypeScript frontend, and releases are frequent: v3.87.4 shipped on 9 October 2026.
The editor accepts Markdown and HTML and exports to Word, PDF and Markdown. Content can be pulled in from a URL, a sitemap, an RSS feed or offline files. Readers get AI search and question answering over your own documents, an embeddable web widget, and chatbots for DingTalk, Feishu, WeCom and Discord. You bring the chat model: the docs recommend deepseek-v4-flash and say embedding and reranking models are built in, so a first install only needs a chat model configured.
Free does not mean unlimited, which earlier versions of this page did not say. The open-source edition allows one wiki site, 300 documents per wiki and one administrator. Professional, aimed at public wikis, is listed at ¥5,000 a year and raises that to ten sites, 10,000 documents and 20 administrators, adding multi-language sites, custom AI prompts and advanced SEO. Commercial, aimed at internal wikis, adds single sign-on, visitor permissions, watermarks, version history, an MCP server and API tokens; it and Enterprise are sold by quote. Paid editions are closed source. Chaitin also sells a hosted version through its Baizhi Cloud.
The caveats are language and licence. The docs, the support community and the bundled chat integrations assume a Chinese-speaking team, and multi-language sites are a paid feature. AGPL-3.0 matters too: modify the open-source edition and run it as a service for others, and you must publish your changes. The company behind it is well capitalised, having raised more than ¥1 billion in 2024 and ¥500 million in January 2026.
Compiled by RECATOOLS from public sources: vendor documentation and pricing, code repositories and published reporting.
About this listing
This entry was compiled from publicly available data including PandaWiki's official website, press releases, documentation, and reputable third-party publications. RECATOOLS is not affiliated with PandaWiki unless explicitly stated.
Third-party AI tools update their pricing, features, availability, and policies frequently. Information here may be outdated by the time you read this — we make reasonable efforts to keep listings current, but cannot guarantee absolute accuracy.
For the latest details, please refer to PandaWiki directly →
Spotted something out of date? Suggest an update →
Related reading
Alternatives to PandaWiki
More in Productivity
