ai-memory vs karpathy-llm-wiki
Markdown knowledge tools for coding agents compared
ai-memory records coding-agent sessions and handoffs in a self-hosted, git-backed wiki. karpathy-llm-wiki guides an Agent Skills-compatible agent to turn source material into curated Markdown pages and answer with citations. The main difference is session memory and cross-agent continuity versus a source-based research and knowledge workflow.
ai-memory: Share Long-Term Memory Across Coding Agents
A self-hosted Rust service that captures coding-agent activity into a searchable, git-backed Markdown wiki and hands work between agents. It suits developers and teams who want portable project memory without requiring LLM calls.
karpathy-llm-wiki: Build a Citation-Based LLM Knowledge Wiki
An Agent Skills-compatible workflow that turns source material into a maintained Markdown wiki for coding agents. Use it when you want durable, cross-linked knowledge and cited answers rather than retrieving raw documents for every question.
| ai-memory | karpathy-llm-wiki | |
|---|---|---|
| Language | Rust | Python |
| License | MIT | MIT |
| Stars | 9.1k | 2.4k |
| Forks | 632 | 286 |
| Last analyzed | Oct 10, 2026 | Oct 3, 2026 |
Key differences
- ai-memory captures prompts, tool calls, and session boundaries; karpathy-llm-wiki ingests documents and compiles them into wiki articles.
- ai-memory provides a Rust server and CLI with search and handoffs; karpathy-llm-wiki is a Python skill and workflow, not a hosted wiki application or general-purpose retrieval service.
- ai-memory supports full-text, entity, and link search, with optional vector search; karpathy-llm-wiki emphasizes ordinary file and text search over vector or graph tooling.
- ai-memory is self-hosted and can share project knowledge across machines and teammates; karpathy-llm-wiki relies on an Agent Skills-compatible coding agent and Markdown directories managed by the user.
- Both projects use the MIT license; ai-memory is written in Rust, while karpathy-llm-wiki is written in Python.
- The supplied figures list ai-memory at 8.8k stars and 608 forks, and karpathy-llm-wiki at 2.4k stars and 286 forks.
Choose ai-memory if you…
- need to preserve coding-agent sessions, decisions, and handoffs across agents or machines.
- want a self-hosted, git-backed project wiki with basic capture, search, and handoffs that do not require LLM API calls.
- prefer project knowledge stored as inspectable and editable Markdown files, with optional vector search or LLM features.
Choose karpathy-llm-wiki if you…
- want an Agent Skills-compatible workflow to turn papers, web pages, and other source material into curated Markdown articles.
- need an agent to answer questions from wiki pages with citations and help maintain links and indexes.
- prefer curated pages and ordinary text search over a standalone retrieval service or vector search tooling.
Compare the numbers side by side, or add more projects
This comparison is generated with AI from the OSRepos analyses of both projects. Always check each project's repository and documentation before choosing.