Matt Wolf

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

The market seems to think Anthropic just won the AI war against OpenAI. But what's the real story?

Everyone is talking about Anthropic's massive $965B valuation filing compared to OpenAI's $852B. And if we look at the business fundamentals, Anthropic brought the receipts:
💰 $30B vs $25B in annualized revenue
🤝 70% win rate for massive enterprise clients
📈 Projected profitability by 2028 (OpenAI isn't expected until 2030)
🇻🇦 And the cherry on top: a acknowledgment on AI safety from the Pope himself

BUT... OpenAI is still the undefeated king of the consumer market with a staggering 905 MILLION weekly active users for ChatGPT.

The real battle isn't just about valuation. It’s Enterprise Revenue vs. Consumer Growth. Claude is winning on precision and nuance, while ChatGPT is winning on convenience and brand dominance.

Which team are you on? Let me know in the comments!

#ChatGPT #OpenAI #Claude #Anthropic #Ainews

The future is here, and it’s ready to take your crap 😂

This is a self driving toilet called Xiaoban made by Chinese company Yueban that just debuted at a Shanghai aged-care expo.

You just summon it to your bedside and it has a built-in bidet. When you’re done, it autonomously drives to your regular bathroom, dumps the waste into the toilet, scrubs its own bowl, and docks itself to recharge and refill.

Obviously this was made for folks with mobility issues, and it’s pretty cool that tech has come this far to be able to help people in this way!

At the same time, I can’t help but imagine some people using this just for fun or out of laziness.

#AI #Robotics #FutureTech

This is the biggest mistake I see people making with AI.

We’ve all fallen into the trap of trying to use AI for EVERYTHING. But the difference between an AI novice and a power user is knowing when to step back.

The goal isn't to automate your entire life. It's to know which tasks AI should automate, which ones it should help with, and which ones are still better left to humans.

That’s exactly what I’ll be covering in my webinar with @TeachableTV as part of their AI Academy.

We'll go through a practical framework for deciding what to automate (and what not to), real-world AI workflows you can start using immediately, common automation mistakes to avoid, and how to use AI as a collaborator without sacrificing your own creativity and judgment.

Register at the link in the comments. And while you're there, let me know: what's something you've tried to automate with AI that completely flopped?

#teachablepartner #AIforbusiness #contentcreation #Ai #aiagents

How to generate motion graphics using AI:

Step 1: Choose your AI assistant
These frameworks act as plugins or "skills" that work directly inside AI coding assistants. In the video, I used Claude Code, but they also work with VS Code, GitHub Copilot, Codex, Cowork, and more.

Step 2: Find the framework you want to use
To make this work, you'll need a programmable video framework like Remotion or Hyperframes. You can find ones like Remotion or Hyperframes (the ones I use in this video) by simply searching for them on GitHub or in a search engine.

Step 3: Install it
You don’t need to be a developer to set this up. I literally went into the Claude Code prompt box, pasted the GitHub URL for the framework, and typed: "Install this for me, please." That is actually how easy it is.

Step 4: Write your prompt
Just describe the exact animation you want. The AI writes the underlying code and automatically renders out a clean MP4 file that you can drop directly into your video edit.

If you need professional looking graphics but don't want to spend hours keyframing in After Effects, this workflow is going to save you an enormous amount of time.

✅ SAVE this reel so you can try out this workflow next time you need it!

#motiongraphics #videoediting #AI #AIhack #contentcreator

Stop losing money on separate AI subscriptions. Get ChatGPT, Claude, Gemini, and 200+ models in one place for one price with Nexos. Get 50% off with my code: https://nexos.ai/wolf

Sakana AI just released a model that beats Fable on some benchmarks, ties it on others, and loses on a few, and we put it to the test ourselves this week.

This week's AI news isn't huge in volume, but it's massive in stakes. I break down Sakana's new Fugu model, an orchestrator system that routes your prompts across multiple AI models instead of being just one model itself, and test it head to head against Fable on a couple of real build challenges. I also cover Anthropic's new Claude Tag feature, which lets you tag Claude directly inside Slack like a coworker, and the news that the US government is now getting directly involved in how OpenAI releases new models.

Plus a full rapid fire on Seed Dance 2.5, Krea 2 going open weights, a new searchable database of AI training data from The Atlantic, OpenAI's own inference chip news, and a wave of new Meta Ray-Ban styles.

If you want to stay looped in on AI news every single week, make sure to subscribe. I post a news breakdown every Friday and tool tutorials midweek.

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

#AINews #SakanaFugu #claudeai

Time Stamps:
00:00 Sakana Fugu: A New AI Model Router
02:00 Testing Fugu vs. Fable (Setup & Cost)
04:24 Fugu Test Results: Game & Website Clones
08:21 Nexos
10:03 Anthropic's Claude Tag
12:19 OpenAI, GPT 5.6 & Government Oversight
14:10 Rapid Fire: Seed Dance, Krea 2 & More
18:34 Final Thoughts & Outro

Thanks to Nexos for sponsoring this video.

Here's the truth about that MIT AI study everyone's been referencing.

Yes, leaning on AI too early or too completely can make you "dumber." But the full study reveals a much more nuanced reality: human skill + AI performs better than human skill alone.

When you already know how to think critically, AI becomes a superpower. When you use it to replace all critical thinking, well... you may lose the ability to think for yourself 🤷‍♂️

What's another AI claim you keep hearing about that's not entirely true, or more complicated than people are saying?

#AI #science

64 8

9 free skills and plug-ins that make Claude Code, Codex, and other AI coding agents dramatically more useful, and yes, all of them are free.
Skills and plug-ins are reusable instruction files that give your AI agent a repeatable behavior instead of starting from scratch every time. In this video I break down 9 of my favorites, install them live, and put several of them head to head so you can see exactly what changes.

I cover GStack (turns Claude Code into a full virtual engineering team), Stop Slop (strips AI tells out of your writing), Graphify (turns any codebase or knowledge base into a queryable graph your agent can use as memory), Understand Anything (visual onboarding maps for any codebase), Last 30 Days (real time sentiment research across Reddit, X, YouTube, and more), Anthropic's Front End Design skill and the Taste skill (head to head design tests), and ReMotion vs HyperFrames (AI generated animations from a single prompt).

Every skill works across Claude Code, Codex, Cursor, OpenClaw, Hermes, VS Code, and GitHub Copilot. Links to every single one below.

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

Skills and plug-ins mentioned:
GStack: https://github.com/garrytan/gstack
Stop Slop: https://github.com/hardikpandya/stop-slop
Graphify: https://github.com/safishamsi/graphify
Understand Anything: https://github.com/Egonex-AI/Understand-Anything
Last 30 Days: https://github.com/mvanhorn/last30days-skill
Front End Design (Anthropic): https://www.skills.sh/anthropics/skills/frontend-design
Taste Skill: https://github.com/Leonxlnx/taste-skill
ReMotion: https://github.com/remotion-dev/remotion
HyperFrames: https://github.com/heygen-com/hyperframes

If you want more breakdowns like this, let me know in the comments and I'll keep them coming.

Let’s work together!
- Brand, sponsorship & business inquiries: mattwolfe@smoothmedia.co

Timestamps:
00:00 Why Skills Matter
01:05 Skills vs Plugins Explained
02:45 GStack: Your AI Engineering Team
08:14 Stop Slop: Cleaner AI Writing
09:25 Graphify: Knowledge Graphs as Memory
14:14 Understand Anything: Visual Code Maps
17:55 Newsletter Quick Plug
18:39 Last 30 Days: Real Time Sentiment Research
20:55 Front End Design vs. Taste Skill
24:46 Animation Showdown: ReMotion vs. HyperFrames
28:07 Final Thoughts and Outro

#ClaudeCode #AIAgents #FreeAITools

2.2K 94

I messed up.

Everyone has been talking about the drama that went down around Claude Fable and Mythos getting shut down, including myself. But I got distracted from the REAL story here, which is just how crazy good these models were and the amazing things people were able to build with them.

These are just a few examples, but it's insane that people built full scale projects and business ideas and NONE of them were engineers with massive dev budgets. They were just regular people with a good idea.

What used to require a full development team and months of work now takes a single afternoon and a good AI model. We are watching in real time a massive shift in WHO gets to build things.

But of course, the party was cut short because Anthropic was ordered to suspend foreign access to Fable 5 over security concerns, literally describing their own tech as "too powerful."

And I get why that sounds terrifying. But look at what people actually did with it in the short window they had: they solved problems, built businesses, and created cool things to share with the world. That’s what happens when you put genuinely useful tech into the hands of the people.

#Claude #Anthropic #AI #Entrepreneurship #vibecoding

532 19

Codex can now watch your screen, learn what repetitive tasks you need done, and then do them for you.

OpenAI just dropped a new feature called "Record and Replay" and it’s the ultimate cheat code for repetitive tasks.

First, you record yourself doing a task once and the AI will watch your screen, understand your clicks, and turn that task into a brand-new "skill" that it saves. Then, the next time you need to do this task, just drop whatever context you need into the chat box and tell it to use the skill it saved.

What is the FIRST repetitive task you would let this AI copy for you? Let me know in the comments👇

#OpenAI #AI #ProductivityHacks #AInews #chatgpt

1.6K 72

Did PewDiePie actually build a ChatGPT competitor?

Felix dropped $41,000 on an insane local AI rig and built his own open-source AI workspace with zero prior coding knowledge. While the internet is calling it a ChatGPT competitor, it’s actually something much more important: a completely private, self-hosted interface that keeps your data OUT of the hands of Big Tech companies.

It’s one of the greatest examples of what open-source AI and "vibe coding" makes possible today.

👇 What’s the most questionable AI claim you’ve seen lately? Drop it below and I’ll try to debunk it in the next video!

#AI #PewDiePie #ChatGPT #vibecoding

1.3K 36

Here's The AI News You Probably Missed This Week. Learn more about how Box AI can unlock key insights for your business here:
https://www.box.com/ai?utm_source=youtube&utm_medium=paidinfluencer&utm_theme=icm&utm_campaign=FY27_Q2_MattWolfe_June19

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

Resources From Today's Video:
Fable/Mythos Suspension: https://www.anthropic.com/news/fable-mythos-access
Anthropic Export Controls: https://www.politico.com/news/2026/06/13/inside-the-whirlwind-24-hours-that-led-the-white-house-to-slap-export-controls-on-anthropic-00961519
Jassy Security Concerns: https://x.com/steph_palazzolo/status/2065830580135051306
Transparent AI Letter: https://freefable.org/
Mythos/Fable Access Returning: https://www.koreajoongangdaily.com/business/anthropic-confident-of-reenabling-mythos-fable-5-access-in-coming-days-executive/12727522
GLM-5.2 Launch: https://z.ai/blog/glm-5.2
Facebook AI Tools: https://about.fb.com/news/2026/06/new-ai-tools-to-help-you-make-things-happen-on-facebook/
Midjourney New Era: https://www.midjourney.com/medical/blogpost
Whole-Body Ultrasounds: https://x.com/hankgreen/status/2067471250159448305
Codex Record Replay: https://x.com/OpenAIDevs/status/2067681320281723113
Claude Design Branding: https://claude.com/blog/claude-design-stays-on-brand-for-daily-work
Self-Improving Agent Memory: https://www.perplexity.ai/hub/blog/self-improving-memory-for-agents
Palmier Video Editor: https://x.com/Marcos12345rico/status/2067264877463261400
Adobe AI Assistant: https://techcrunch.com/2026/06/18/adobe-adds-its-ai-assistant-to-premiere-illustrator-and-indesign/
Ask Ad Manager: https://blog.google/products/admanager/ask-ad-manager/
Americans AI Views: https://www.pewresearch.org/internet/2026/06/17/americans-and-ai-2026-chatbots-smart-devices-and-views-on-impact/
Self-Driving Toilet: https://www.theverge.com/tech/952441/yueban-xiaoban-self-driving-autonomous-toilet
Let’s work together!
- Brand, sponsorship & business inquiries: mattwolfe@smoothmedia.co

#AINews #AITools #ArtificialIntelligence

Time Stamps:
0:00 Intro
0:09 Fable / Mythos Banned
6:38 Box AI
8:19 GLM-5.2
15:07 Facebook AI
17:04 Midjourney Scanner
24:06 OpenAI Record & Replay
25:16 Claude Design Update
25:57 Perplexity Brain
27:19 Palmier
28:34 Adobe AI Updates
28:57 Google Ask Ad Manager
29:38 People use but hate AI
33:00 Self-Driving Toilet
34:14 Final Thoughts
35:29 WTF?

2.8K 350