fast-rlm

A minimal implementation of Recursive Language Models (RLMs) using Deno and Pyodide.

Watch the full video on YouTube RLM Tutorial

What are RLMs

RLMs are an inference technique where an LLM interacts with arbitrarily long prompts through an external REPL. The LLM can write code to explore, decompose, and transform the prompt. It can recursively invoke sub-agents to complete smaller subtasks. Crucially, sub-agent responses are not automatically loaded into the parent agent's context — they are returned as symbols or variables inside the parent's REPL.

Support

If you find this helpful, consider supporting on Patreon — it hosts all code, projects, slides, and write-ups from the YouTube channel.

---

Install

pip install fast-rlm

Requirements

• Python 3.10+
• Deno 2+
  • macOS/Linux: curl -fsSL https://deno.land/install.sh | sh
  • Windows (npm): npm install -g deno
• (Optional) Bun — only needed for the TUI log viewer

Environment Variables

Set your LLM API key before running:

export RLM_MODEL_API_KEY=sk-or-...

Variable	Description	Default
RLM_MODEL_API_KEY	API key for your LLM provider	—
RLM_MODEL_BASE_URL	OpenAI-compatible base URL	https://openrouter.ai/api/v1

By default, fast-rlm uses OpenRouter. You can point it at any OpenAI-compatible API by setting RLM_MODEL_BASE_URL.

---

Quick Start

import fast_rlm

result = fast_rlm.run("Generate 50 fruits and count number of r")
print(result["results"])
print(result["usage"])

Configuration

from fast_rlm import run, RLMConfig

config = RLMConfig.default()
config.primary_agent = "minimax/minimax-m2.5"
config.sub_agent = "minimax/minimax-m2.5"
config.max_depth = 5
config.max_money_spent = 2.0

result = run(
    "Count the r's in 50 fruit names",
    prefix="r_count",
    config=config,
)

All config fields:

Field	Type	Default	Description
primary_agent	str	z-ai/glm-5	Model for the root agent
sub_agent	str	minimax/minimax-m2.5	Model for child subagents
max_depth	int	3	Max recursive subagent depth
max_calls_per_subagent	int	20	Max LLM calls per subagent
truncate_len	int	2000	Output chars shown to the LLM per step
max_money_spent	float	1.0	Hard budget cap in USD

Log Viewer

Every run saves a .jsonl log file to logs/.

# Print stats (no extra dependencies)
fast-rlm-log logs/run_xxx.jsonl

# Interactive TUI viewer (requires bun)
fast-rlm-log logs/run_xxx.jsonl --tui

---

Development (from source)

1. Install Deno

Windows (npm):

npm install -g deno

macOS / Linux:

curl -fsSL https://deno.land/install.sh | sh

Then add Deno to your PATH:

export DENO_INSTALL="$HOME/.deno"
export PATH="$DENO_INSTALL/bin:$PATH"

2. Install Bun (for the log viewer)

curl -fsSL https://bun.sh/install | bash
cd tui_log_viewer && bun install

3. API Key Setup

Set your key in .env or .envrc:

export RLM_MODEL_API_KEY=sk-or-...

4. Configuration

Edit rlm_config.yaml at the project root:

max_calls_per_subagent: 20
max_depth: 3
truncate_len: 2000
primary_agent: "z-ai/glm-5"
sub_agent: "minimax/minimax-m2.5"
max_money_spent: 1.0

5. Running

# Run the example
deno task test_counting_r

# Run the subagent directly
echo "What is 2+2?" | deno task subagent

# View logs
./viewlog logs/<logfile>.jsonl

6. Benchmarks

uv sync --extra benchmarks
uv run benchmarks/oolong_synth_benchmark.py
uv run benchmarks/longbench_benchmark.py