System Requirements

1. Install Ollama

2. Choosing GLM 5.1 model

3. OpenClaw

4. OpenClaw Control Panel (Control UI)

5. Connecting automatic web search

6. Connecting the messaging platform

7. Using with Claude Code and Codex (Bonus)

About the GLM-5.1 Model: Why It Matters OpenClaw Challenge Submission 🦞 This is a submission for the OpenClaw Writing Challenge In this guide, I'll walk you through installing three tools step by step that together give you a free personal AI assistant running right on your computer.

No subscriptions, no monthly fees = completely free. Here's what we'll install: Ollama:A program that lets you run AI models directly on your computer. Think of it as an "engine" that powers artificial intelligence locally. Works on macOS, Windows, and Linux. GLM-5.1:An AI model from Chinese company Z.AI (formerly Zhipu AI). Released on April 7, 2026, it's one of the best open-source models in the world. On the SWE-Bench Pro benchmark (a coding test), it scored 58.4 points, more than GPT-5.4 and Claude Opus 4.6. And it's completely free under the MIT license. OpenClaw:An AI agent that transforms a regular language model into a full-featured assistant. It can send messages via WhatsApp, Telegram, Slack, Discord, search the web, work with files, write code, and automate tasks. OpenClaw launches through Ollama with a single command. Before installing, make sure your computer meets these requirements: Minimum (for cloud models via Ollama Cloud): Recommended (for local models): Important:If you don't have a powerful GPU, don't worry! GLM-5.1 is available as a cloud model through Ollama (glm-5.1:cloud), and it runs fast without any hardware requirements on your end. Ollama is the foundation of the entire system. We start here.Install Ollama by following the link to the official website. Paste the command into the terminal or download the app. Now Ollama is installed. Just type ollama and the program will launch, then select the option “Chat with a model”. Next, select the desired model, in my case GLM 5.1.You'll be able to chat with it right in the terminal, ask questions, request code, analyze texts, and more. Important:To use cloud models, you need to sign in to an Ollama account. If you don't have one yet, Ollama will prompt you to create one when you first launch a cloud model. It's free. Now we are connected via the cloud version of GLM 5.1. Next, going down a little further on the same site, we will install OpenClaw using the command (if it is not already installed): Launch with configuration before starting:If you want to configure OpenClaw first (choose a model, connect messengers, etc.): After launching OpenClaw, you can open the web control panel in your browser. Go to: In the Control UI you'll find: At the top of the panel, you can switch between channels and models. Next, we follow the documentation and enable automatic web search. OpenClaw can search for information on the internet. If you're using a cloud model through Ollama, web search is enabled automatically.For local models, you need to install the plugin: Now you can connect messengers that will make it easier to communicate with your agent. One of OpenClaw's key features is the ability to chat with the AI agent through familiar messengers. You send a message on Telegram and get a response from AI. Channel setup is done through the Control Panel (Control UI) or via the command: Select the “Channels” section and follow the instructions to connect the messenger you need. Link WhatsApp, Telegram, Slack, Discord, or iMessage with this command: For example, to connect a Telegram account, you will need to: Next we add our bot token in the terminal, and we're all set. Ollama also lets you launch other AI development tools. As shown on the GLM-5.1 page on Ollama, there are ready-made commands: Claude Code with GLM-5.1: OpenCode with GLM-5.1: This lets you harness the power of GLM-5.1 through different programming interfaces. GLM-5.1 is a next-generation model from Z.AI (formerly Zhipu AI, a Tsinghua University spinoff). Key facts: GLM-5.1's key feature is its ability to work autonomously for up to 8 hours, revising its strategy and finding new approaches to problem-solving.While other models “run out of steam” after their initial attempts, GLM-5.1 continues improving results through hundreds of iterations. Three commands. Zero subscriptions. Your own AI agent that works for you, not someone else's server. The future is here and it's free. 🔖 Bookmark this – you'll need it again.

🔔 Follow my profile, so you catch guides like this first. Templates let you quickly answer FAQs or store snippets for re-use. Hide child comments as well For further actions, you may consider blocking this person and/or reporting abuse