Back to directory

Karpathy LLM Wiki Plugin for Obsidian

This Obsidian plugin implements Andrej Karpathy's LLM Wiki idea directly inside the editor, turning notes and PDFs into a connected, queryable knowledge base with entity pages, concept pages and graph-powered question answering.

Karpathy LLM Wiki Plugin for Obsidian

This plugin turns an Obsidian vault into a linked, LLM-powered knowledge base, built as a one-click plugin rather than a separate CLI tool, desktop app or coding-agent skill the way other reimplementations of Karpathy's LLM Wiki idea are delivered. It runs inside Obsidian itself, using the built-in Graph View, ribbons and command palette, and needs no separate backend, database or embedding model.

It supports more than 16 LLM providers, including Anthropic, OpenAI, Bedrock, Gemini, DeepSeek, Qwen and Ollama, ingests Markdown, PDF, image and Office documents natively, and retrieves related notes with a zero-embedding graph algorithm, personalized PageRank with Monte Carlo sampling, over the vault's wiki-link graph rather than chunk-based vector search. Its interface and generated wiki output are available in 11 languages.

It is local-first and does not require a backend server, and is written in TypeScript under the Apache-2.0 license.

More in Agent memory

See all 450
  • A long technical guide to AI agent memory: why stateless LLMs need it, the different memory architectures from vector stores to knowledge graphs, and how Cognee implements one as an open-source memory layer.

  • A local, file-based long-term memory layer for a personal AI that never summarizes conversations, keeping every line with a timestamp and searching by time first and words second, rather than retrieving only similar-seeming chunks.

  • NEOTH

    deepwiki.com

    A local-first personal AI daemon in Rust with five memory tiers, consent-gated tools, a WASM plugin sandbox, multi-provider model routing and signed audit logs, aimed at people who want both a daily assistant and a serious coding helper.