ChatGLM

How to use GLM 5.2. This is a brand new open model that scores near the top of the pack while costing a fraction of the big paid AI tools. In this video, I'll show you how to use it step by step, starting free in your browser and then running it inside your own tools like Claude Code. We start at chat.z.ai, where anyone can use GLM 5.2 for free with no setup at all. I'll show you real tasks like writing a launch announcement, pulling themes out of customer feedback, and summarizing a long document using the model's huge context window. Then we level up by creating an API key, adding a little funds, and pointing Claude Code at GLM 5.2 so you can keep your exact same workflow for just pennies. I'll also be honest about where it fits, including why you almost certainly don't need to run this giant model locally on your own machine. Z.AI Chat (free GLM 5.2): https://chat.z.ai GLM Coding Plan / API Keys: https://z.ai/manage-apikey/apikey-list Code Block: "env": { "ANTHROPIC_AUTH_TOKEN": "your_new_zai_api_key", "ANTHROPIC_BASE_URL": "https://api.z.ai/api/anthropic", "ANTHROPIC_DEFAULT_OPUS_MODEL": "glm-5.2", "ANTHROPIC_DEFAULT_SONNET_MODEL": "glm-5.2", "API_TIMEOUT_MS": "3000000" } Timestamps: 0:00 Intro 0:30 What GLM 5.2 Is and Why It Matters 1:18 Getting Started Free at chat.z.ai 1:57 First Task: Writing a Launch Announcement 2:36 Deep Think Settings (High vs Max) 3:04 Pulling Themes From Customer Feedback 3:37 Summarizing a Long Document With the Large Context Window 4:08 Agent Mode and the AI Slide Maker 4:48 Moving Beyond the Browser 5:01 Creating Your API Key 5:54 The Paid Side and Adding Funds 6:26 Setting Up GLM 5.2 in Claude Code 8:33 Building a One-Page Landing Site 10:14 Using the Same Key in Other Tools 10:44 Should You Run It Locally? 11:19 Wrap Up

Z.ai's GLM-5.2 is a 1 million token, MIT-licensed open weight model that costs a fraction of frontier AI prices, and I put it through real tests to show you exactly where it holds up and where it doesn't. Also join my newsletter: https://futuretools.io/newsletter I cover the three ways to actually use it (hosted web app, API and agent harness, or self-hosted if you have the infrastructure), build a webpage with it live, run it inside Cursor to create self-improving automation skills from my own meeting notes, generate a Remotion animation comparing GLM-5.2 against Opus 4.6, GPT-5.5, and Gemini 3.5, and show you a tool from Inference.net that lets you mirror production traffic to GLM-5.2 risk-free before fully committing. If your workflows are long, code-heavy, document-heavy, or burning through tokens faster than you'd like, this model is worth serious attention. It's not beating Claude or GPT across the board, but for the right use cases the cost difference alone changes the math. All links mentioned are below. Resources: https://gptzero.me/ https://z.ai/subscribe Discover More: 🛠️ Explore AI Tools & News: https://futuretools.io/ 📰 Weekly Newsletter: https://futuretools.io/newsletter Socials: ❌ Twiter/X: https://x.com/mreflow 🖼️ Instagram: https://instagram.com/mr.eflow 🧵 Threads: https://www.threads.net/@mr.eflow 🟦 LinkedIn: https://www.linkedin.com/in/matt-wolfe-30841712/ 👍 Facebook: https://www.facebook.com/mattrwolfe Let’s work together! - Brand, sponsorship & business inquiries: mattwolfe@smoothmedia.co Chapters: 00:00 What Is GLM-5.2 and Why It Matters 02:18 Three Ways to Access It 05:19 Newsletter Quick Plug 05:57 Live Tests: Webpage, Logic & Ethics 11:56 BuseyBench: SVG Challenge 13:51 Using GLM-5.2 in Cursor 15:03 MegaBonk Clone Test 17:22 Building a Chrome Extension & Cleaning Downloads 20:12 "Improve Your Matt": Self-Improving AI Skills 23:36 Remotion Animation Test 25:10 Inference.net: Risk-Free Model Switching 26:28 Final Verdict and Outro #ZAI #GLM52 #OpenSourceAI