We are going to install OpenClaw on a Raspberry Pi 4 and use it as a bot controlled by Telegram today.
First on the Pi, update the system and install the basic dependencies.
sudo apt update && sudo apt upgrade -y sudo apt install -y git curl build-essential sudo timedatectl set-timezone Europe/Madrid
Install Node.js 24.
curl -fsSL https://deb.nodesource.com/setup_24.x | sudo -E bash - sudo apt install -y nodejs node --version
Install OpenClaw.
curl -fsSL https://openclaw.ai/install.sh | bash
Update PATH without rebooting.
source ~/.bashrc
<
p Run the onboarding. When asked, select QuickStart, then filter by groq and select llama-3.3-70b-versatile, choose DuckDuckGo for searches, say No to all optional API keys, No to skills and No to hooks.
openclaw onboard --install-daemon
Once the onboarding is complete, configure the Telegram channel.
To do this, first create a bot in Telegram by talking to @BotFather, run /newbot, give it a name and username, and save the token it gives you.
Luego abre el configurador.
openclaw configure
Choose Channels, then Configure/Link, select Telegram and paste the BotFather token when asked.
Then start the gateway.
openclaw gateway
Write a message to your bot in Telegram. The first time it will ask for matching approval. In another terminal execute this to approve the access.
openclaw pairing list telegram openclaw pairing approve telegram CODE
And that’s it, the assistant responds in Telegram using llama-3.3-70b-versatile from Groq on the Pi.
And now configure Groq to be able to connect via llm:
OpenCLAW Configure
Select Model, it will ask for your Groq API key, paste the one you have. You can get it for free on console.groq.com if you don’t have it handy.
Groq does not appear directly in the list. Select Custom Provider and press Enter. There you put the Groq data manually:
Select OpenAI-compatible (it’s already marked) and press Enter. Groq is compatible with OpenAI format.
To use Ollama, we use this:
Base URL: https://dominio-ollama.com/v1
API Key: ollama
Model: llama3.1:8b
