https://github.com/zircote/rlm-rs

Rust CLI implementing the Recursive Language Model (RLM) pattern for Claude Code. Process documents 100x larger than context windows through intelligent chunking, SQLite persistence, and recursive sub-LLM orchestration.
https://github.com/zircote/rlm-rs

ai-tools chunking claude claude-code cli command-line context-window devtools document-processing llm mit-license mmap rayon recursive-language-model rlm rust rust-2024 semantic-chunking sqlite text-processing

Last synced: 10 days ago
JSON representation

Rust CLI implementing the Recursive Language Model (RLM) pattern for Claude Code. Process documents 100x larger than context windows through intelligent chunking, SQLite persistence, and recursive sub-LLM orchestration.

• Host: GitHub
• URL: https://github.com/zircote/rlm-rs
• Owner: zircote
• License: mit
• Created: 2026-01-19T13:33:21.000Z (about 1 month ago)
• Default Branch: main
• Last Pushed: 2026-01-19T21:09:04.000Z (30 days ago)
• Last Synced: 2026-01-19T22:46:07.385Z (30 days ago)
• Topics: ai-tools, chunking, claude, claude-code, cli, command-line, context-window, devtools, document-processing, llm, mit-license, mmap, rayon, recursive-language-model, rlm, rust, rust-2024, semantic-chunking, sqlite, text-processing
• Language: Rust
• Size: 159 KB
• Stars: 0
• Watchers: 0
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: .github/readme-infographic.svg
  • Contributing: CONTRIBUTING.md
  • License: LICENSE
  • Codeowners: CODEOWNERS

Awesome Lists containing this project

README

          

  

    

      

      

    

    

      

      

    

    

      

      

    

    

      

      

    

  

  

  

  

  

    How RLM Works

  

  

  

    

    Large Document

    

    

      

      

      

      

      

      

      

      

      

      

    

    100x context size

  

  

  

    

    

    chunk

  

  

  

    Chunks

    

    

    chunk_0.txt

    

    chunk_1.txt

    

    chunk_2.txt

    

    chunk_n.txt

  

  

  

    

    

    process

  

  

  

    

    Sub-LLM

    (Haiku)

    

    

      

      analyze

    

    

      

      extract

    

    

      

      summarize

    

  

  

  

    

    

    aggregate

  

  

  

    

    Root LLM

    (Opus/Sonnet)

    

    

      

      

    

    Final Result

  

  

  

    

    

    

      

      Semantic Chunking

    

    

      

      SQLite Persistence

    

    

      

      Memory-Mapped I/O

    

    

      

      Parallel Processing