🚀 Getting Started¶
Welcome to RLM Code, the Research Playground and Evaluation OS for Recursive Language Model (RLM) agentic systems. RLM Code provides a unified TUI-based development environment for building, benchmarking, and optimizing agent workflows through slash commands and natural language.
🧪 What is RLM Code?¶
RLM Code implements the Recursive Language Model paradigm from the 2025 "Recursive Language Models" paper. It extends the paper's concepts with:
- 🧠 Context-as-variable: Context is stored as a REPL variable rather than in the token window, enabling unbounded output and token-efficient processing
- 🔁 Deep recursion: Support for recursion depth > 1, exceeding the paper's original limitation
- 🔀 Multi-paradigm execution: Pure RLM, CodeAct, and Traditional paradigms with side-by-side comparison
- 📊 Pluggable observability: MLflow, OpenTelemetry, LangSmith, LangFuse, and Logfire integrations
- 📦 Sandbox runtimes: Local, Docker, Apple Container, Modal, E2B, and Daytona execution environments
🎯 Problem Focus¶
RLM (the method) addresses long-context reasoning. RLM Code is the tooling layer for researchers who want to implement, evaluate, and operate that workflow.
RLM Code is optimized for workflows where:
- Context is too large to fit comfortably in one prompt.
- You need programmatic inspection and decomposition instead of full-context prompt injection.
- You want to compare recursive symbolic execution against harness-style and direct baselines under the same benchmark suite.
For detailed mode behavior and neutral tradeoff guidance, see Execution Patterns.
📚 Where to Go Next¶
| Guide | Description |
|---|---|
| 🧭 Start Here (Simple) | Plain-language onboarding: what this is, what to install, and safe first run |
| 📦 Installation | System requirements, package installation, optional dependencies, and verification |
| ⚡ Quick Start | Launch the TUI, connect a model, run your first benchmark, explore the Research tab |
| 🧑🔬 Researcher Onboarding | Researcher-first workflows and complete command handbook |
| 💻 CLI Reference | Complete reference for the entry point and all 50+ slash commands |
| ⚙️ Configuration | Full rlm_config.yaml schema, environment variables, and ConfigManager API |
⚡ Quick Overview¶
# Install
pip install rlm-code
# Launch the unified TUI
rlm-code
# Connect to a model and run a benchmark
/connect anthropic claude-opus-4-6
/rlm bench preset=dspy_quick
/rlm bench compare candidate=latest baseline=previous
🆕 First Time?
Start with the 📦 Installation guide to set up your environment, then follow the ⚡ Quick Start for a hands-on walkthrough.
🖥️ Unified TUI
RLM Code ships a single TUI with 5 tabs: 🔁 RLM, 📁 Files, 📋 Details, ⚡ Shell, and 🔬 Research. Use rlm-code to launch, and press Ctrl+5 to access the Research tab for experiment tracking, benchmarks, and session replay.