RuVector vs cortex-mem
Persistent agent memory projects compared
RuVector and cortex-mem are Rust projects that help AI applications retain and retrieve information across sessions. RuVector focuses on vector search and local retrieval across a broad set of components, while cortex-mem combines file-based memory, LLM extraction, and semantic search in a more integrated system that depends on external services.
RuVector: Build Agent Memory with Rust and ONNX
RuVector is a Rust-based vector search and persistent agent-memory toolkit, with ONNX embeddings for local semantic retrieval. It offers Rust, Node.js, and browser/WASM paths, plus separate graph, learning, and MCP components.
cortex-mem: Give AI Agents Persistent Long-Term Memory
Cortex Memory is a Rust framework for storing, extracting, and searching long-term memories for AI agents and applications. It combines durable file-based storage with semantic search, and offers CLI, REST API, and MCP access.
| RuVector | cortex-mem | |
|---|---|---|
| Language | Rust | Rust |
| License | MIT | MIT |
| Stars | 4.6k | 311 |
| Forks | 613 | 24 |
| Last analyzed | Oct 19, 2026 | Oct 3, 2026 |
Key differences
- RuVector offers local ONNX embeddings and supports external embedding providers; cortex-mem uses configured LLM and embedding services and Qdrant for semantic search.
- RuVector spans separate vector, graph, learning, browser/WASM, and other package surfaces; cortex-mem organizes Markdown memories into abstracts, overviews, and detailed content in a virtual filesystem.
- RuVector can be used with a local persistent store through its Node.js path; running the full cortex-mem system requires Qdrant and compatible model services.
- RuVector’s components require explicit integration, and its considerations identify incomplete or limited areas in some runtime and persistence capabilities; cortex-mem presents extraction, indexing, and memory updates as part of its system.
- Both projects are Rust-based and MIT-licensed; RuVector has 4.6k stars and 613 forks, while cortex-mem has 311 stars and 24 forks.
Choose RuVector if you…
- need local semantic retrieval with ONNX embeddings or want to supply your own embeddings in a Rust application.
- want vector-search components across Rust, Node.js, or browser/WASM paths.
- are prepared to select and integrate separate graph, learning, or MCP components as needed.
Choose cortex-mem if you…
- want a layered Markdown memory system with summaries and detailed content.
- need conversation memory extraction, automated indexing, and cleanup in one system.
- want CLI, REST API, MCP, or Rust integration, and can run Qdrant and compatible model services.
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.