I Failed 5 Times Setting Up MoltBot Before I Finally Got It Working - Complete Beginner Guide With Every Step Explained 2026
I Failed 5 Times Setting Up MoltBot Before I Finally Got It Working
The complete beginner guide with every single step explained - no coding experience needed
20 min read • Step by Step • Troubleshooting IncludedReal talk. The first time I tried to install MoltBot I broke something on my computer and had to reinstall the whole operating system.
Second time I got stuck on a permission error for three hours.
Third time the bot connected but it would not respond to any messages.
Fourth time it worked for 10 minutes then crashed and never came back.
Fifth time I almost gave up completely.
If any of this sounds familiar, keep reading. I am going to save you from the same frustration.
😤 Why Is It So Hard?
Most tutorials online assume you already know things. They say stuff like "just clone the repo and run the setup wizard" like everyone knows what that means.
Or they skip steps because it seems obvious to them. But if you are not a developer, nothing is obvious.
And the official documentation is written by developers for developers. Great if you speak that language. Useless if you do not.
The Real Problems I Faced (And How I Fixed Them)
Let me share every single problem I encountered so you can avoid them:
Ollama Not Starting Properly
What happened: I installed Ollama but when I tried to use it, nothing worked. The command just hung there doing nothing.
✅ Solution: Ollama needs to run as a background service, not just be installed. On Linux run: sudo systemctl start ollama and then sudo systemctl enable ollama to make it start automatically.
The Environment File Confusion
What happened: I edited the .env.example file directly and nothing worked. Spent hours wondering why my settings were being ignored.
✅ Solution: You must COPY the example file and rename it. Run: cp .env.example .env first, then edit the new .env file. The example file is just a template.
Telegram Bot Not Responding
What happened: I created the bot with BotFather and got the token, but when I messaged the bot it just ignored me completely.
✅ Solution: You need to START a conversation with your bot first by clicking the Start button in Telegram. Also make sure MoltBot is actually running. Check with: pm2 status or look for the process.
Firewall Blocking Connections
What happened: Everything seemed configured correctly but MoltBot could not connect to Ollama. No error message, just silence.
✅ Solution: Your firewall might be blocking local connections. On Linux, allow the Ollama port: sudo ufw allow 11434. On Windows, add an exception in Windows Firewall for port 11434.
Out of Memory Crashes
What happened: The model would start loading then my computer would freeze or the process would just die. No clear error message.
✅ Solution: Your model is too big for your RAM. If you have 8GB use llama3.2:3b. If you have 16GB use qwen2.5:7b. Also close Chrome and other memory-hungry apps.
💡 What Finally Worked
I found a guide that was written for normal people. Not developers. It explained every single step with screenshots. Every command you need to type. Every setting you need to change. Every common error and how to fix it.
I followed it step by step and in about 2 hours I had everything running. The same thing that took me weeks of failed attempts before.
The Complete Setup Process (Actually Explained)
Here is the full process broken down into steps that actually make sense:
📦 Phase 1: Preparing Your Computer
Choose Your Operating System
MoltBot works on Windows, Mac, and Linux. But Linux is recommended because it is more stable for running 24/7.
Best options for beginners:
• Ubuntu Desktop 22.04 - Easiest Linux for beginners
• Ubuntu Server 22.04 - If you do not need a graphical interface
• Windows 11 - Works but needs more RAM
• macOS - Works great on M1/M2/M3 Macs
Check Your Hardware
Before you start, make sure your computer meets these requirements:
| RAM | 8GB minimum (16GB recommended) |
| Storage | 20GB free space (SSD preferred) |
| Processor | Any dual-core from 2015+ |
| GPU | Not required (optional speed boost) |
⚙️ Phase 2: Installing Ollama
Download and Install Ollama
On Linux or Mac - Open Terminal and paste this command:
On Windows - Go to ollama.com/download and download the Windows installer. Run it like any normal program.
⚠️ Wait for it to finish! The installation might take 2-5 minutes. Do not close the terminal window.
Start Ollama Service
On Linux - Run these commands to start Ollama and make it run automatically:
sudo systemctl enable ollama
On Mac - Ollama starts automatically. You will see a llama icon in your menu bar.
On Windows - Ollama runs automatically after installation. Look for the icon in your system tray.
Download an AI Model
Now download a model based on your RAM:
ollama pull llama3.2:3b
# If you have 16GB RAM:
ollama pull qwen2.5:7b
# If you have 32GB+ RAM:
ollama pull qwen2.5:32b
ℹ️ This takes time! Models are 2-20GB depending on size. Could take 10-30 minutes on slow internet. Be patient.
Test Ollama
Before continuing, make sure Ollama works. Run this command:
If you get a response, Ollama is working. If not, check the troubleshooting section below.
🤖 Phase 3: Installing MoltBot
Install Node.js (Required)
MoltBot needs Node.js to run. Install it first:
curl -fsSL https://deb.nodesource.com/setup_20.x | sudo -E bash -
sudo apt-get install -y nodejs
# On Mac:
brew install node
On Windows: Download from nodejs.org and install normally.
Download MoltBot
Clone the MoltBot repository from GitHub:
cd moltbot
⚠️ Don't have git? Install it first: sudo apt install git on Linux or download from git-scm.com
Create Configuration File (IMPORTANT!)
This is where most people mess up. You MUST copy the example file:
🚫 DO NOT edit .env.example directly! Always copy it to .env first. This is the #1 mistake beginners make.
Configure MoltBot for Ollama
Open the .env file in a text editor and add these settings:
OLLAMA_HOST=http://localhost:11434
DEFAULT_MODEL=qwen2.5:7b
DEFAULT_PROVIDER=ollama
# Your Telegram Bot Token (get from @BotFather)
TELEGRAM_BOT_TOKEN=your_token_here
To edit on Linux: nano .env then Ctrl+X to save.
📱 Phase 4: Setting Up Telegram
Create a Telegram Bot
Open Telegram and search for @BotFather. Start a chat and follow these steps:
- Send the command:
/newbot - Choose a name for your bot (example: "My AI Assistant")
- Choose a username ending in "bot" (example: "myai_helper_bot")
- BotFather will give you a token that looks like:
123456789:ABCdefGHIjklMNOpqrSTUvwxYZ - Copy this token and paste it in your .env file
Start Your Bot
Search for your bot username in Telegram and click START. This is required before the bot can send you messages.
✅ Important: You MUST click Start first. Many people skip this step and wonder why the bot does not respond.
🚀 Phase 5: Launch and Test
Install Dependencies and Run
Install required packages and start MoltBot:
npm start
You should see messages indicating MoltBot is connected and waiting for messages.
Test Your AI Assistant
Open Telegram and send a message to your bot. Try something simple like:
"Hello! Can you tell me a joke?"
🎉 If you get a response, congratulations! Your personal AI assistant is now running on your own hardware.
Common Errors and Quick Fixes
If something goes wrong, check these common issues:
| Error Message | Solution |
|---|---|
Connection refused |
Ollama is not running. Start it with: sudo systemctl start ollama |
Model not found |
Download the model first: ollama pull qwen2.5:7b |
Out of memory |
Use a smaller model. Try llama3.2:3b instead. |
Telegram timeout |
Check your bot token is correct. Get a new one from @BotFather. |
Permission denied |
Add sudo before the command, or fix folder permissions. |
npm not found |
Node.js is not installed. Install it first. |
💻 My Current Setup (Working for 4 Months)
Hardware
Old ThinkPad T460
RAM
16GB (upgraded)
OS
Ubuntu Server 24.04
Model
Qwen 2.5 7B
Uptime
4 months, zero issues
Monthly Cost
~$3 electricity
Still Struggling With Setup?
I spent weeks figuring all this out through trial and error. You do not have to. Everything is documented in one place with screenshots for every single step.
- ✓ 231 pages of step-by-step instructions
- ✓ Screenshots of every single screen
- ✓ Troubleshooting for 50+ common errors
- ✓ Copy-paste configurations included
- ✓ Works on Windows, Mac, and Linux
This book saved me from weeks of frustration. It can save you too.
Final Words
Setting up MoltBot and Ollama is not hard once you know the right steps. The problem is most guides skip important details or assume you already know things.
I failed five times before getting it right. Now I have an AI assistant that has been running 24/7 for four months without a single issue. The investment of time was absolutely worth it.
Your own AI assistant is closer than you think. Follow this guide step by step and you will get there.
✅ What You Learned Today
The 5 most common setup problems and how to fix them • Complete installation process for Ollama • How to configure MoltBot correctly • Setting up Telegram bot integration • Troubleshooting common errors • My working setup after 4 months
Made for everyone who struggled with the setup like I did
MoltBot and ClawdBot are open source projects. This blog is not officially affiliated.