Matthew Berman's banner

Matthew Berman

@MatthewBerman • 127,534 subscribers

The future is bright. Get Ahead - https://t.co/s5tbuuNdjE Loop Skill - https://t.co/1B5X282n8S Learn Loops - https://t.co/8t0scQUngV

Shorts

Ok this is insane. Scene from the original Halo game. Includes what seems like 100% accurate sound effects.

Ok this is insane. Scene from the original Halo game. Includes what seems like 100% accurate sound effects.

1,331,774 просмотров

But can Sora write code? "have someone explain how to write python code on a black board that outputs "hello world""

But can Sora write code? "have someone explain how to write python code on a black board that outputs "hello world""

377,845 просмотров

Major AI breakthrough: Diffusion Large Language Models are here! They're 10x faster and 10x cheaper than traditional LLMs. Here's everything you need to know:

Major AI breakthrough: Diffusion Large Language Models are here! They're 10x faster and 10x cheaper than traditional LLMs. Here's everything you need to know:

429,113 просмотров

You can use Michael Jackson's likeness...for now.

You can use Michael Jackson's likeness...for now.

99,674 просмотров

Sora 2 Pro is even better at coding.

Sora 2 Pro is even better at coding.

98,128 просмотров

I did the thing to the robot NVIDIA AI PC Check out behind the scenes content on my IG: matthewberman_ai

I did the thing to the robot NVIDIA AI PC Check out behind the scenes content on my IG: matthewberman_ai

32,533 просмотров

Videos

Anya Rossi

sweetdream.ai

SweetDream.ai•Sponsored•Livecam

Watch Anya Live

Anya is streaming live right now! Join her private show and enjoy exclusive content.

Exclusive private shows

1.2k viewers online

Private Show

Join now for exclusive access

Free preview available • Premium content

If you can only choose one subscription (ChatGPT or Claude), the choice has become obvious.

If you can only choose one subscription (ChatGPT or Claude), the choice has become obvious.

119,459 просмотров • 3 дней назад

I’ve been using GPT-5.6 Sol internally for the past two months, I've spent probably 25+ billion tokens. Here’s my review and comparison to Fable 5: > Let's start with the analogy because everyone seems to be giving theirs - GPT-5.6 is likely the last version of the GPT-5 training run series. It's kind of like an athlete at their peak. Through years of experience in the game, they've become the most reliable player and has the highest game IQ. But, there's no more room to grow. Fable on the other hand, being essentially the first version of a new training run, is the first round draft pick rookie. Raw talent mixed with the energy only a young person would have results in some incredible plays we didn't think possible, but also mistakes due to lack of experience. But that rookie will only improve and likely will be better than the veteran ever was because it's a new game and a new era. > GPT-5.6 is genuinely better at long, sustained work. With /goal, I've had it running complex projects for days with almost no intervention. It built a Minecraft-style game, kept adding features and mobs after the core game worked, and only stopped because I stopped the run. I never felt as though I had to jump in and guide it back to the right path. > It keeps finding useful work when you give it a concrete finish line. I had it recreate Excel with a loop. It inspected the real desktop excel app with Computer Use, comparing that against its own build, and closing the gaps. I stopped it after six days after it had built an incredible amount of functionality. > It's faster than other models in two different ways. The raw generation speed is higher, something OpenAI has been putting effort into. But it also takes a shorter path to solutions. It wanders less, changes less code, and generally knows how to get things done directly. In daily use, it feels about 2-3x times faster than Fable. That's my impression, not a controlled benchmark. The difference is large enough that I notice it constantly. > It works well across a wide range of tasks. I use it for one-line edits, quick questions, browser chores, and multi-day builds without changing my prompting style. Speaking of browser control, its the best ever I've used. To the point where I actually use it often. If a task lives on a website, GPT-5.6 usually opens the browser and does it there instead of asking for an API key or forcing everything through the terminal. When I switched back to GPT-5.5, it went straight to the command line even when the browser was clearly the better tool. > And it can handle real browser work, not just toy demos. During a data import, I had it monitor Supabase and resize instances as the load changed. It stayed on the dashboard, adjusted capacity, and checked the result without an API or a custom script. > I also gave it a full Google Workspace migration. It moved Forward Future from to preserved the old aliases, and configured MX, SPF, and DKIM. Before a consequential save, it stopped, explained exactly what would change, and waited for confirmation. > The reasoning setting matters a lot. Light is good for questions and small edits. High and Extra High are the sweet spots for serious work. Ultra usually takes longer than the extra thinking is worth and burns tokens. > I love that 5.6 is split into 3 sizes. Not only can you control speed and cost that way, but you still also have the thinking effort setting for each of them. Very precise controls. I just wish Codex automatically routed my prompts for me. > Its personality is blunt and a little bland. Claude feels warmer and more natural to talk to. GPT-5.6 is more clinical, but I like that for work. It gives me enough explanation and rarely pads the answer. I usually have to ask Fable to explain things more simply and/or more concise. > Its front-end taste has improved, but the default is predictable. Left alone, it turns websites into PowerPoint decks with huge statements and hard section breaks. The good news is that it takes design direction well and can revise without destroying the parts that already work. > It still makes confident mistakes. I asked it to rebuild parts of a system, and it told me the job was finished. Later, I found out it wasn't. Bits of its internal process also leak into the answer occasionally. > Claude Fable is more naturally autonomous on large, open-ended projects. GPT-5.6 is easier to reach for. I don't need to invent a huge project to justify using it. It works just as well for a small edit or browser chore. > GPT-5.6 is also cheaper. Sol costs $5 per million input tokens and $30 per million output tokens. Fable costs $10 and $50. Cached input is cheaper too. Still, cost per finished task matters more than cost per token. > GPT-5.6 isn't the best at everything, and it still needs supervision. But it generates faster, wanders less, works at almost any scale, and wastes less of my time. It's the model I have the most confidence in to get the job done right the first time. I put together a full breakdown with all the tests, prompts, and examples on a site. You can read it here:

I’ve been using GPT-5.6 Sol internally for the past two months, I've spent probably 25+ billion tokens. Here’s my review and comparison to Fable 5: > Let's start with the analogy because everyone seems to be giving theirs - GPT-5.6 is likely the last version of the GPT-5 training run series. It's kind of like an athlete at their peak. Through years of experience in the game, they've become the most reliable player and has the highest game IQ. But, there's no more room to grow. Fable on the other hand, being essentially the first version of a new training run, is the first round draft pick rookie. Raw talent mixed with the energy only a young person would have results in some incredible plays we didn't think possible, but also mistakes due to lack of experience. But that rookie will only improve and likely will be better than the veteran ever was because it's a new game and a new era. > GPT-5.6 is genuinely better at long, sustained work. With /goal, I've had it running complex projects for days with almost no intervention. It built a Minecraft-style game, kept adding features and mobs after the core game worked, and only stopped because I stopped the run. I never felt as though I had to jump in and guide it back to the right path. > It keeps finding useful work when you give it a concrete finish line. I had it recreate Excel with a loop. It inspected the real desktop excel app with Computer Use, comparing that against its own build, and closing the gaps. I stopped it after six days after it had built an incredible amount of functionality. > It's faster than other models in two different ways. The raw generation speed is higher, something OpenAI has been putting effort into. But it also takes a shorter path to solutions. It wanders less, changes less code, and generally knows how to get things done directly. In daily use, it feels about 2-3x times faster than Fable. That's my impression, not a controlled benchmark. The difference is large enough that I notice it constantly. > It works well across a wide range of tasks. I use it for one-line edits, quick questions, browser chores, and multi-day builds without changing my prompting style. Speaking of browser control, its the best ever I've used. To the point where I actually use it often. If a task lives on a website, GPT-5.6 usually opens the browser and does it there instead of asking for an API key or forcing everything through the terminal. When I switched back to GPT-5.5, it went straight to the command line even when the browser was clearly the better tool. > And it can handle real browser work, not just toy demos. During a data import, I had it monitor Supabase and resize instances as the load changed. It stayed on the dashboard, adjusted capacity, and checked the result without an API or a custom script. > I also gave it a full Google Workspace migration. It moved Forward Future from to preserved the old aliases, and configured MX, SPF, and DKIM. Before a consequential save, it stopped, explained exactly what would change, and waited for confirmation. > The reasoning setting matters a lot. Light is good for questions and small edits. High and Extra High are the sweet spots for serious work. Ultra usually takes longer than the extra thinking is worth and burns tokens. > I love that 5.6 is split into 3 sizes. Not only can you control speed and cost that way, but you still also have the thinking effort setting for each of them. Very precise controls. I just wish Codex automatically routed my prompts for me. > Its personality is blunt and a little bland. Claude feels warmer and more natural to talk to. GPT-5.6 is more clinical, but I like that for work. It gives me enough explanation and rarely pads the answer. I usually have to ask Fable to explain things more simply and/or more concise. > Its front-end taste has improved, but the default is predictable. Left alone, it turns websites into PowerPoint decks with huge statements and hard section breaks. The good news is that it takes design direction well and can revise without destroying the parts that already work. > It still makes confident mistakes. I asked it to rebuild parts of a system, and it told me the job was finished. Later, I found out it wasn't. Bits of its internal process also leak into the answer occasionally. > Claude Fable is more naturally autonomous on large, open-ended projects. GPT-5.6 is easier to reach for. I don't need to invent a huge project to justify using it. It works just as well for a small edit or browser chore. > GPT-5.6 is also cheaper. Sol costs $5 per million input tokens and $30 per million output tokens. Fable costs $10 and $50. Cached input is cheaper too. Still, cost per finished task matters more than cost per token. > GPT-5.6 isn't the best at everything, and it still needs supervision. But it generates faster, wanders less, works at almost any scale, and wastes less of my time. It's the model I have the most confidence in to get the job done right the first time. I put together a full breakdown with all the tests, prompts, and examples on a site. You can read it here:

183,716 просмотров • 9 дней назад

I've spent 2.54 BILLION tokens perfecting OpenClaw. The use cases I discovered have changed the way I live and work. ...and now I'm sharing them with the world. Here are 21 use cases I use daily: 0:00 Intro 0:50 What is OpenClaw? 1:35 MD Files 2:14 Memory System 3:55 CRM System 7:19 Fathom Pipeline 9:18 Meeting to Action Items 10:46 Knowledge Base System 13:51 X Ingestion Pipeline 14:31 Business Advisory Council 16:13 Security Council 18:21 Social Media Tracking 19:18 Video Idea Pipeline 21:40 Daily Briefing Flow 22:23 Three Councils 22:57 Automation Schedule 24:15 Security Layers 26:09 Databases and Backups 28:00 Video/Image Gen 29:14 Self Updates 29:56 Usage & Cost Tracking 30:15 Prompt Engineering 31:15 Developer Infrastructure 32:06 Food Journal

I've spent 2.54 BILLION tokens perfecting OpenClaw. The use cases I discovered have changed the way I live and work. ...and now I'm sharing them with the world. Here are 21 use cases I use daily: 0:00 Intro 0:50 What is OpenClaw? 1:35 MD Files 2:14 Memory System 3:55 CRM System 7:19 Fathom Pipeline 9:18 Meeting to Action Items 10:46 Knowledge Base System 13:51 X Ingestion Pipeline 14:31 Business Advisory Council 16:13 Security Council 18:21 Social Media Tracking 19:18 Video Idea Pipeline 21:40 Daily Briefing Flow 22:23 Three Councils 22:57 Automation Schedule 24:15 Security Layers 26:09 Databases and Backups 28:00 Video/Image Gen 29:14 Self Updates 29:56 Usage & Cost Tracking 30:15 Prompt Engineering 31:15 Developer Infrastructure 32:06 Food Journal

3,370,531 просмотров • 5 месяцев назад

Use Fable + GPT-5.5 for peak performance and lowest quota usage.

Use Fable + GPT-5.5 for peak performance and lowest quota usage.

180,809 просмотров • 11 дней назад

I'm one of the most advanced users of OpenClaw. OpenClaw + GPT5.3 Codex + Opus 4.6 has been the trifecta that changed everything. I made a video going over everything I'm doing with these tools. Learn these tools, stay ahead. Watch this video right now. 0:00 Intro 1:02 Overview 4:17 Sponsor 5:12 Personal CRM 7:11 Knowledge Base 8:30 Video Idea Pipeline 11:09 Twitter/X Search 12:47 Analytics Tracker 13:33 Data Review 15:34 HubSpot 16:13 Humanizer 16:52 Image/Video Generation 18:22 To-Do List 19:37 Usage Tracker (Saves Money) 20:45 Services 21:25 Automations 22:42 Backup 23:30 Memory 24:06 Building OpenClaw 25:22 Updating Files

I'm one of the most advanced users of OpenClaw. OpenClaw + GPT5.3 Codex + Opus 4.6 has been the trifecta that changed everything. I made a video going over everything I'm doing with these tools. Learn these tools, stay ahead. Watch this video right now. 0:00 Intro 1:02 Overview 4:17 Sponsor 5:12 Personal CRM 7:11 Knowledge Base 8:30 Video Idea Pipeline 11:09 Twitter/X Search 12:47 Analytics Tracker 13:33 Data Review 15:34 HubSpot 16:13 Humanizer 16:52 Image/Video Generation 18:22 To-Do List 19:37 Usage Tracker (Saves Money) 20:45 Services 21:25 Automations 22:42 Backup 23:30 Memory 24:06 Building OpenClaw 25:22 Updating Files

2,480,676 просмотров • 5 месяцев назад

Dario ruined it for everyone

Dario ruined it for everyone

285,606 просмотров • 22 дней назад

"We think that with AI we can replace all of our Jr developers in our company" AWS CEO Matt Garman: "That's the dumbest thing I've ever heard"

"We think that with AI we can replace all of our Jr developers in our company" AWS CEO Matt Garman: "That's the dumbest thing I've ever heard"

698,037 просмотров • 1 месяц назад

5 BILLION tokens later, OpenClaw is now my company's operating system. I discovered things most people never will. (PS I solved the Anthropic OAuth loophole.) Here’s exactly how it works. 0:00 Intro 0:16 Email Management 5:20 Sponsor 7:02 Inbox Pipeline 9:05 Multiple Prompt Versions (HUGE) 12:28 MD File Breakdown 14:12 Telegram Groups 14:51 CRM System 17:25 Meeting Intelligence 18:45 Knowledge Base 20:51 Content Pipeline 21:53 Security (HUGE) 24:49 Cron Jobs 26:05 Memory 27:55 Notification Batching 28:59 Financial Tracking 29:40 Usage & Cost Tracking 31:01 Full Logging Infrastructure 31:52 OAuth Loophole (HUGE) 32:55 Separating Personal/Work 34:29 Errors & Self-Improvement 35:59 Cost Savings 37:09 Backup & Recovery 37:52 Health Pipeline 38:30 Bee Memory

5 BILLION tokens later, OpenClaw is now my company's operating system. I discovered things most people never will. (PS I solved the Anthropic OAuth loophole.) Here’s exactly how it works. 0:00 Intro 0:16 Email Management 5:20 Sponsor 7:02 Inbox Pipeline 9:05 Multiple Prompt Versions (HUGE) 12:28 MD File Breakdown 14:12 Telegram Groups 14:51 CRM System 17:25 Meeting Intelligence 18:45 Knowledge Base 20:51 Content Pipeline 21:53 Security (HUGE) 24:49 Cron Jobs 26:05 Memory 27:55 Notification Batching 28:59 Financial Tracking 29:40 Usage & Cost Tracking 31:01 Full Logging Infrastructure 31:52 OAuth Loophole (HUGE) 32:55 Separating Personal/Work 34:29 Errors & Self-Improvement 35:59 Cost Savings 37:09 Backup & Recovery 37:52 Health Pipeline 38:30 Bee Memory

1,138,917 просмотров • 4 месяцев назад

Moltbots/Clawdbots now have their own social network (moltbook) and it's wild. This is the first time I'm a little scared... You need to watch this.

Moltbots/Clawdbots now have their own social network (moltbook) and it's wild. This is the first time I'm a little scared... You need to watch this.

1,178,222 просмотров • 5 месяцев назад

wtf is a loop?

wtf is a loop?

252,815 просмотров • 1 месяц назад

Every AI lab is starving for compute. Except Google. I spoke with Thomas Kurian, Google Cloud's CEO, to understand why Google doesn't just hoard compute before AGI, their relationship with Anthropic, and that viral tweet about Google's engineering culture. Watch now: 0:00 – Intro 0:42 – Google's Insane Compute Capacity 03:17 – TPU Monetization 05:24 – Why Google Doesn't Hoard Compute? 08:02 – Datacenter Buildout 15:01 - Does AGI Mean Job Displacement? 17:55 - NVIDIA vs TPU (Total Cost of Ownership) 23:25 - 8th Gen TPU 24:32 - Training vs. Inference 30:53 - Google's "Extreme Co-design" 35:01 - Working with Anthropic 37:46 - Serving Mythos-sized Models (10T) 41:42 - Google engineering culture (Steve Yagge tweet) 48:27 - Cybersecurity 51:50 - What keeps Thomas up at night?

Every AI lab is starving for compute. Except Google. I spoke with Thomas Kurian, Google Cloud's CEO, to understand why Google doesn't just hoard compute before AGI, their relationship with Anthropic, and that viral tweet about Google's engineering culture. Watch now: 0:00 – Intro 0:42 – Google's Insane Compute Capacity 03:17 – TPU Monetization 05:24 – Why Google Doesn't Hoard Compute? 08:02 – Datacenter Buildout 15:01 - Does AGI Mean Job Displacement? 17:55 - NVIDIA vs TPU (Total Cost of Ownership) 23:25 - 8th Gen TPU 24:32 - Training vs. Inference 30:53 - Google's "Extreme Co-design" 35:01 - Working with Anthropic 37:46 - Serving Mythos-sized Models (10T) 41:42 - Google engineering culture (Steve Yagge tweet) 48:27 - Cybersecurity 51:50 - What keeps Thomas up at night?

507,872 просмотров • 2 месяцев назад

American open source AI is in trouble. China is eating our lunch. This is a bigger problem than people realize.

American open source AI is in trouble. China is eating our lunch. This is a bigger problem than people realize.

451,194 просмотров • 2 месяцев назад

LOOPS - everything you need to know:

LOOPS - everything you need to know:

148,763 просмотров • 29 дней назад

Sundar Pichai (Sundar Pichai), Google CEO, on: 🔹Race to AGI 🔹Agents 🔹AI & Information Diet 🔹Open Source 🔹Cybersecurity 🔹US vs China Intro AI agents replacing parts of the internet AI agents deciding our information diet Will agents kill the “raw internet”? AI cyberattacks and Google’s defense strategy Should dangerous cyber AI models be held back? The threshold for releasing powerful AI Why Google hasn't open-sourced a large model The broken business model with open-source AI American companies using Chinese AI The risk of building on China’s AI ecosystem Why Google cares so much about cheap, fast AI models Sundar on self-improving AI and the race to AGI Google’s compute situation Google has more AI demand than compute Google’s biggest compute bottleneck

Sundar Pichai (Sundar Pichai), Google CEO, on: 🔹Race to AGI 🔹Agents 🔹AI & Information Diet 🔹Open Source 🔹Cybersecurity 🔹US vs China Intro AI agents replacing parts of the internet AI agents deciding our information diet Will agents kill the “raw internet”? AI cyberattacks and Google’s defense strategy Should dangerous cyber AI models be held back? The threshold for releasing powerful AI Why Google hasn't open-sourced a large model The broken business model with open-source AI American companies using Chinese AI The risk of building on China’s AI ecosystem Why Google cares so much about cheap, fast AI models Sundar on self-improving AI and the race to AGI Google’s compute situation Google has more AI demand than compute Google’s biggest compute bottleneck

300,861 просмотров • 1 месяц назад

Dylan Patel says we aren't ready for what's coming... Round 2 with Dylan Patel 1:13 - Dylan's predictions 7:47 - Anthropic vs DoW 15:08 - War Claude 22:00 - How happiness in society works 31:31 - Knowledge work is cooked 38:22 - Is SaaS dead? 45:18 - New Media landscape 48:16 - White collar bloodbath 52:38 - Open Source is Losing 1:04:45 - Chinese AI Distillation Attacks 1:09:52 - Closed Source VS Open Source 1:19:43 - Microsoft CEO is coping 1:26:55 - Who wins the ASI race?

Dylan Patel says we aren't ready for what's coming... Round 2 with Dylan Patel 1:13 - Dylan's predictions 7:47 - Anthropic vs DoW 15:08 - War Claude 22:00 - How happiness in society works 31:31 - Knowledge work is cooked 38:22 - Is SaaS dead? 45:18 - New Media landscape 48:16 - White collar bloodbath 52:38 - Open Source is Losing 1:04:45 - Chinese AI Distillation Attacks 1:09:52 - Closed Source VS Open Source 1:19:43 - Microsoft CEO is coping 1:26:55 - Who wins the ASI race?

643,549 просмотров • 4 месяцев назад

Anthropic discovered J-space and it's a big deal. 🔸Where models do deep (invisible) thinking 🔸No lying in the J-space (alignment solution) 🔸We can modify the J-space to change AI's thoughts Full breakdown:

Anthropic discovered J-space and it's a big deal. 🔸Where models do deep (invisible) thinking 🔸No lying in the J-space (alignment solution) 🔸We can modify the J-space to change AI's thoughts Full breakdown:

48,681 просмотров • 10 дней назад

Demis says he wants to see a Western open source AI stack and that we’re losing to China. He also says Google doesn’t have enough compute to build two frontier (open and closed) models, which is why Gemma is a smaller family of models. Watch this incredible clip. Shout out Y Combinator and Garry Tan for the fantastic interview.

Demis says he wants to see a Western open source AI stack and that we’re losing to China. He also says Google doesn’t have enough compute to build two frontier (open and closed) models, which is why Gemma is a smaller family of models. Watch this incredible clip. Shout out Y Combinator and Garry Tan for the fantastic interview.

296,424 просмотров • 2 месяцев назад

Rivian CEO RJ Scaringe: "The next 10 years will be the most important period in human history." The full interview: 2:45 Rivian’s big 2026 moment 4:23 The origin of Rivian 6:25 The pivot that changed Rivian 7:21 Rivian’s core mission 9:06 Obsessing over details 9:57 Why R2 matters 12:07 Cutting cost, keeping quality 15:21 One brain, thousands of decisions 18:56 Rivian’s software advantage 19:02 Autonomy and the physical world 20:28 The AI shift in self-driving 23:26 Rivian’s autonomy roadmap 25:26 Training AI from real driving 28:41 R2 as a data machine 29:45 Vision vs LiDAR 35:43 Safety and corner cases 37:35 Fewer cars or more driving? 40:03 Robotaxis vs car ownership 42:21 RJ’s robotics thesis beyond the humanoid hype 47:59 How to raise kids for an unrecognizable future 50:40 The timeline that should worry everyone

Rivian CEO RJ Scaringe: "The next 10 years will be the most important period in human history." The full interview: 2:45 Rivian’s big 2026 moment 4:23 The origin of Rivian 6:25 The pivot that changed Rivian 7:21 Rivian’s core mission 9:06 Obsessing over details 9:57 Why R2 matters 12:07 Cutting cost, keeping quality 15:21 One brain, thousands of decisions 18:56 Rivian’s software advantage 19:02 Autonomy and the physical world 20:28 The AI shift in self-driving 23:26 Rivian’s autonomy roadmap 25:26 Training AI from real driving 28:41 R2 as a data machine 29:45 Vision vs LiDAR 35:43 Safety and corner cases 37:35 Fewer cars or more driving? 40:03 Robotaxis vs car ownership 42:21 RJ’s robotics thesis beyond the humanoid hype 47:59 How to raise kids for an unrecognizable future 50:40 The timeline that should worry everyone

397,688 просмотров • 4 месяцев назад

Google's Genie 3 is capable of generating full 3D worlds. But not only that, they are FULLY INTERACTIVE. Here are the most incredible demos they shared:

Google's Genie 3 is capable of generating full 3D worlds. But not only that, they are FULLY INTERACTIVE. Here are the most incredible demos they shared:

893,706 просмотров • 11 месяцев назад

This is it. Clawdbot has gone viral and for good reason. Here's a full explanation of what it is, how to use it, and yes, even its shortcomings. You need to watch this.

This is it. Clawdbot has gone viral and for good reason. Here's a full explanation of what it is, how to use it, and yes, even its shortcomings. You need to watch this.

427,584 просмотров • 5 месяцев назад