# LLM Wiki

> Updated 2026-09-12 · type: tool · category: productivity-knowledge · status: active · rev 1

LLM Wiki compiles your documents into a persistent cited wiki once, instead of re-reading them on every question like standard RAG.

- Open source: yes (GPL-3.0)
- Self-hostable: yes
- Pricing model: free
- Best for: Operators who keep re-asking an LLM about the same playbooks, call notes and research, and want compiled, persistent, citable knowledge rather than retrieval from scratch.
- Last verified: 2026-09-12

- **Canonical:** https://gtmstacker.com/registry/tool/llm-wiki/
- **Source:** [github · nashsu/llm_wiki](https://github.com/nashsu/llm_wiki)
- **Tags:** productivity-knowledge, knowledge-base, rag, second-brain, self-hostable, desktop
- **Repository:** https://github.com/nashsu/llm_wiki

## Is LLM Wiki open source?

Yes, LLM Wiki is open source under the GPL-3.0 license.

## How much does LLM Wiki cost?

LLM Wiki is free to use.

## Can I self-host LLM Wiki?

Yes, LLM Wiki can be self-hosted (the source is available under the GPL-3.0 license).

## Alternatives & related

- [Raggy](https://gtmstacker.com/registry/tool/raggy/)


---

Open-source desktop app (GPL-3.0, 19.1k★; Tauri, TypeScript + Rust) that compiles your documents into a persistent, interconnected wiki instead of re-reading them per query like standard RAG: two-step chain-of-thought ingest, a knowledge graph with community detection, cited multi-conversation chat, web clipper, and Obsidian-vault compatibility.

## Provenance

- GPL-3.0 independently WebFetch-verified 2026-09-12 (19.1k★, 2.2k forks, 856 commits, active). Surfaced via the 2026-09-12 viral-posts brief (mega-tier author, 437k followers).
- Curated from the GTM Stacker signal registry (2026-09-12 pass: daily pull + viral-posts brief); license independently WebFetch-verified 2026-09-12.

## Why it matters for a GTM stack

The GTM knowledge problem is not retrieval, it is persistence: the same ICP notes, pricing logic and objection handling get re-derived every session. Compile-once into a cited wiki is the right shape for that, and it pairs naturally with a lighter query-time tool (see raggy) rather than replacing it.