Qwen 3.8-Max is Alibaba's 2.4 trillion parameter AI model featuring a sparse mixture of experts architecture that activates only relevant parameters for each task, enabling efficient processing of complex coding and multi-step professional workflows; the model demonstrates practical business applications including landing page copy generation, 14-day content calendar creation, and automated customer support responses, showing how large language models can significantly reduce manual work when properly prompted with clear objectives, voice guidelines, and constraints.
Deep Dive
Prerequisite Knowledge
- No data available.
Where to go next
- No data available.
Deep Dive
New Qwen3.8 Max is INSANE!
Added:NewQuen 3.8 Max is insane. Alibaba just dropped this thing and it might be the second most powerful AI model on the planet right now. Let me say that again because it matters. On July 19th, Alibaba's Quen team put out a new model called Quen 3.8-Max.
It has 2.4 trillion parameters. That's a huge number. Think of parameters like tiny knobs inside the model's brain.
More knobs means it can learn more patterns and handle harder problems.
Alibaba says this model is one of the strongest AI systems out there today and they claim it only trails one model in the world. That model is Anthropic's Claude Fable 5. That's a bold claim.
Alibaba hasn't shown full proof yet. No benchmark table, no model card, no outside testers have run their own checks, but the preview is already live.
You can test it right now through Alibaba's Token Plant and through two of their coding tools Kyoda and Coda Work.
Hey, if we haven't met already, I'm the digital avatar of Julian Goldie, CEO of SEO agency Goldie Agency. Whilst he's helping clients get more leads and customers, I'm here to help you get the latest AI updates. And Julian Goldie reads every comment, so make sure you comment below. Here's why the timing is wild. Just two days before this, a different Chinese company called Moonshot AI released their own giant model called Kimmy K3. That one has 2.8 trillion parameters and it's fully open weight, meaning anyone can download it and use it. Alibaba actually owns a big chunk of Moonshot AI, too. So, you had one company Alibaba invested in dropping a giant open model and then two days later Alibaba's own team dropped their own giant model. It feels like a race and both labs are sprinting. Let's rewind for a second so you understand how fast this is moving. Alibaba's last flagship called Quen 3.7-Max scored 92.4% on a hard science test called GPQA Diamond. It also scored 80.4% on something called SWE-Bench Verified, which tests how well an AI can fix real coding problems. That was already impressive. Now, Qwen 3.8-Max is supposed to be all of that. A year ago, models this size barely existed outside of research labs. Now, Alibaba is shipping something with 2.4 trillion parameters as a live product regular people and businesses can use today.
Here's the part that actually explains why this model might be different. Qwen 3.8-Max uses something called a sparse mixture of experts design. Think of it like a huge hospital full of specialist doctors. When you walk in with a problem, the hospital doesn't wake up every single doctor. It only calls the two or three doctors who actually know how to handle your case. That's what this model does. It has trillions of parameters total, but it only activates a small slice of them for each task.
That's how it can be so big without being painfully slow. It's also Alibaba's first model over 1 trillion parameters that can handle more than just text. It can read images, watch video, and go through documents, too.
Now, here's why any of this actually matters, not just for tech people. This model was built to be strong at two things especially, coding and long multi-step office work. Alibaba is calling that second part professional co-work. Basically, an AI that can sit with you through a long task, not just answer one question and disappear. I want to show you exactly how that plays out. So, let's put it to work on something real, growing the AI Profit Boardroom itself. First one is a landing page. We needed a page that could take someone who's never heard of the AI Profit Boardroom and get them to understand it fast and actually want to join. So, we gave Qwen 3.8-Max this prompt, "Write landing page copy for the AI Profit Boardroom, a community that helps business owners automate their business with AI. Include a headline, three benefit bullet points about saving time and getting more leads, a short trust-building line, and a call to action button. Keep sentences short and simple, third-grade reading level." The model returned a clean headline, three sharp benefit points, and a call to action, ready to hand straight to our designer. That's a real example of how this model can save hours of work, but a tool like Qwen 3.8-Max is only as good as the plan behind it, and that's where most people get stuck staring at a blank screen, not knowing what to even type.
If you're serious about using AI to actually grow your business, that's exactly what we walk you through inside the AI Profit Boardroom. We don't just point you at a tool and leave you guessing. We hand you the exact prompts, the step-by-step playbooks, and a roadmap built around models like Qwen 3.8-Max, so you can start using it the same day for landing pages, content, and customer replies. Whether you run a small shop, an agency, or you're building something on the side, we'll show you how to plug this straight into your week and get real time back. Link is in the comments and the description.
Let's keep going because there's more to unpack here. We wanted a full content calendar to bring new eyes to the AI Profit Boardroom without spending a whole Sunday planning it out by hand.
So, we gave Qwen 3.8-Max this prompt: "Act as a content strategist for the AI Profit Boardroom, a community that teaches business owners AI automation.
Build a 14-day content calendar for YouTube and Instagram. Include a topic, a hook line, and a call to action for each day, all pointed at getting people to join the community." In under a minute, the model gave us a full 14-day calendar with headlines, hooks, and a call to action for every single post.
That's the kind of practical work this model is built for, not flashy demos, actual work that used to eat up your whole afternoon. Here's a third one, and this one's about member support. We wanted a way to answer common questions from people thinking about joining the AI Profit Boardroom without someone typing the same replies over and over all day. We gave Qwen 3.8-Max this prompt: "Read these five common questions people ask before joining the AI Profit Boardroom about coaching calls, the community, and getting support. Write a clear, friendly reply for each one under 50 words, warm but confident tone, and always end with an invite to join. The model returned five clean, ready-to-send replies in under a minute. Each one warm, clear, and ready to paste straight into our inbox. That's the pattern you should notice across every one of these examples. Give the model a clear job, a clear voice, and a clear limit like word count or tone, and it gives you something you can use immediately. That's the actual skill worth learning right now, not just knowing a model exists, but knowing how to talk to it so it gives you something useful on the first try. The skill isn't picking the biggest model every time, it's knowing which tool fits which job, and that's exactly the kind of thing that takes real practice to get right.
That's why we built the AI Profit Boardroom the way we did. Every single week, we run coaching calls where we walk through exactly how to use new models like Qwen 3.8-Max the moment they actually matter for real business use, not months later once everyone else already figured it out. We build the prompts, the workflows, and the step-by-step playbooks so you're never starting from a blank page like I was showing you earlier with the landing page and the content calendar. Inside the community, you'll find other business owners already testing tools like this for content, support, and growth, and you can connect with them directly and swap what's working. If growing your business with AI automation is the goal, that's exactly what we focus on every single day. No fluff, just the tools and the road map to actually use them. Link is in the comments and the description. And if you want the full process behind everything I showed you today, plus the SOPs and over 100 AI use cases just like the landing page, the content calendar, and the support replies you just saw, go join the AI Success Lab. It's completely free. We built it for anyone who wants to actually learn how to use tools like Qwen 3.8-Max the right way without getting lost or overwhelmed. Links are in the comments and the description.
You'll get all the notes and prompts from this exact video sitting inside there waiting for you, plus access to a community of people who are actively building with AI every single day and sharing what's working in real time.
Qwen 3.8-Max is still a preview. The full benchmarks aren't out yet, the license isn't confirmed, and the open weight release hasn't landed, but the direction is clear. Bigger models built for longer tasks moving faster than they were even a year ago. The businesses that win from here aren't the ones with the fanciest tools. They're the ones who actually sit down, test one real task, and build a habit around it. That's the whole game right now.
Related Videos

Expanding Stikbot thumbnails
leopoldshorts
2K views•2023-09-24

Digital Discrimination: Cognitive Bias in Machine Learning
redmonktechevents2974
4K views•2019-12-18

Evolutionary Approach to Clustering by Ujjwal Maulik
ICTStalks
279 views•2019-06-26

Rose Yu "Learning from Large-Scale Spatiotemporal Data"
networkscienceinstitute
2K views•2019-03-04

Stanford Seminar - Generalization through Task Representations with Foundation Models
stanfordonline
4K views•2025-07-14

Satellite-Based Wheat Yield Forecasting using GEE & Transformer Neural Network
gisrsinstitute
634 views•2025-06-15

Paradigm Shifts in Data Processing for the Generative AI Era: Robert Nishihara of Anyscale & Ray.io
GradientFlow
2K views•2025-01-02

How to Build Your Own GenAI-Based Knowledge Management System
2150GmbH
360 views•2025-06-03
Trending

Playstation NO DISC/NO BUY Fight Is Over...
DavidJaffeGames
4K views•2026-07-23

Americans Confused in Australia for 17 Minutes Straight
IWrocker
17K views•2026-07-23

FURIOUS Raskin CORNERS DOJ over Trump DARK PAST!!!!
MeidasTouch
237K views•2026-07-23

NO! They Were The Chosen One!
UFDTech
9K views•2026-07-23