🚀 Gensyn RL-SWARM Node — Ultimate Easy Guide (All Issues Fixed)

⚙️ WATCH VIDEO FULL INSTALLTION PROCESS

👆 Click to watch full video

⚙️ Hardware Requirements

📝 Overview

Your hardware requirements depend on the swarm type and model size you choose.

• For low hardware, use Qwen 0.5B / 1.5B + GSM8K
• For high hardware, use Qwen 7B / 32B / 72B + DAPO-Math 17K

Choose according to the details below:

---

🟩 Small Model (0.5B / 1.5B) — GSM8K Dataset

For users with mid-level hardware.

📌 Minimum VPS Requirements

• 32 GB RAM
• 8 vCPU
• 200 GB Storage
• Best for default (0.5B) training

🖥️ CPU Requirements

• ARM64 or x86_64 CPU
• Minimum 32GB RAM
• ⚠️ Note: Running extra applications during training may cause a crash.

🎮 GPU Requirements (CUDA Supported)

• RTX 3090
• RTX 4090
• A100
• H100

---

🟦 Big Model (7B / 32B / 72B) — DAPO-Math 17K Dataset

For users with powerful hardware.

🔥 Recommended GPUs

• NVIDIA A100 (80GB)
• NVIDIA H100 (80GB)

---

🖥️ Node Setup Guide

🔐 1. Open Your VPS

Connect to your server:

ssh username@ip

🚀 RL-SWARM Installation & Setup Guide

A complete step-by-step guide to run RL-SWARM on Linux (VPS/WSL) or Mac.

---

🛠️ 1. Installation & Setup

Run the following commands step-by-step on your terminal.

1. Install sudo

apt update && apt install -y sudo

2. Install other dependencies

sudo apt update && sudo apt install -y python3 python3-venv python3-pip curl wget screen git lsof nano unzip iproute2 build-essential gcc g++

3. Create a screen session

screen -S gensyn

4. Clone official rl-swarm repo

git clone https://github.com/gensyn-ai/rl-swarm.git && cd rl-swarm

5. Run the swarm

python3 -m venv .venv
. .venv/bin/activate
./run_rl_swarm.sh

• After sometimes, u will see something like this if your running it on Linux system, so here follow the next step
• • Now it will prompt you to login: Follow: 1️⃣ How to Login or access http://localhost:3000/ in VPS? 📶

• Now It will promt Would you like to push models you train in the RL swarm to the Hugging Face Hub? [y/N] Enter N

• Now It will promt >> Enter the name of the model you want to use in huggingface repo/name format, or press [Enter] to use the default model. press Enter & get defalut model:

---❗ If U put model manually then it can be cause of terminated❗--- So better to use Default:

------>>>If u want to select the model then choose between them:

5️⃣ Choose customised model's

 Gensyn/Qwen2.5-0.5B-Instruct

 Qwen/Qwen3-0.6B

 nvidia/AceInstruct-1.5B

 dnotitia/Smoothie-Qwen3-1.7B

 Gensyn/Qwen2.5-1.5B-Instruct

Models (CodeZero):

• Qwen/Qwen2.5-Coder-0.5B-Instruct — Solver role (Recommended Choose This)
• Qwen/Qwen2.5-Coder-1.5B-Instruct — Evaluator (frozen)

Currently, we are running CodeZero on the Gensyn Testnet.

Its Done ✅

NEXT

Detach & Attach from screen

ctrl A D To Detach from screen

• Attach with previous screen:

screen -r gensyn

1️⃣ How to Login or access http://localhost:3000/ in VPS? 📶

• Open a new Terminal and login ur vps

• Allow Incoming connection on VPS

sudo apt install ufw -y
sudo ufw allow 22
sudo ufw allow 3000/tcp

• Enable ufw

sudo ufw enable

• Install cloudflared on the VPS

wget -q https://github.com/cloudflare/cloudflared/releases/latest/download/cloudflared-linux-amd64.deb

sudo dpkg -i cloudflared-linux-amd64.deb

• Check version

cloudflared --version

• Make sure your Node is running on port 3000 in Previous Screen

• Run the tunnel command

cloudflared tunnel --url http://localhost:3000

• Access the Link from your local machine

  

• Now follow Login!

• Done!✅

NEXT

6️⃣ Resolve Bootstraps Error

• RUN THIS CMD : (Must you should be in rl-swarm directory) ❗

cd ~/rl-swarm
pkill -f python
rm -rf /tmp/hivemind-*
rm -rf /tmp/p2pd-*
source .venv/bin/activate
bash run_rl_swarm.sh

👉 Join X for more updates: https://x.com/Hamad__Alpha
💬 If you have any issue: Open an issue on this repo or DM me on X:https://x.com/Hamad__Alpha

Thank you! Best of luck 🚀