memvid-cli

AI memory CLI with crash-safe, single-file storage and semantic search.

Installation

npm install -g memvid-cli

Or use directly with npx:

npx memvid-cli --help

Usage

# Create a new memory file
memvid create my-memory.mv2

# Add documents
memvid put my-memory.mv2 --input document.txt
memvid put my-memory.mv2 --input folder/

# Search
memvid find my-memory.mv2 --query "your search query"

# Ask questions (RAG)
memvid ask my-memory.mv2 --question "What is discussed in the documents?"

# View statistics
memvid stats my-memory.mv2

Features

• Single-file storage - All data in one .mv2 file
• Crash-safe - WAL-based durability
• Semantic search - Vector embeddings with HNSW index
• Lexical search - Full-text search with Tantivy
• Hybrid search - Combine semantic and lexical
• RAG support - Built-in question answering
• Temporal queries - Search by time ranges
• Multi-format - PDF, DOCX, images, audio, and more

Supported Platforms

| Platform | Architecture | |----------|--------------| | macOS | ARM64 (Apple Silicon) | | macOS | x64 (Intel) | | Linux | x64 (glibc) | | Windows | x64 |

Note: Linux ARM64 is not currently supported due to cross-compilation complexity.

Documentation

• Website: https://memvid.com
• Docs: https://docs.memvid.com
• GitHub: https://github.com/memvid/memvid

License

Apache-2.0