Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

I trained models across MacBooks using Apple's AirDrop protocol. grove is a distributed training library for Apple Silicon. Devices discover each other over AWDL, a direct radio link. If there's a shared WiFi network it upgrades to that for speed, otherwise everything goes over the direct link. No router,... show more

Swarnim Jain

2,404 subscribers

667,999 views • 4 months ago •via X (Twitter)

Education Science & Technology

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

built an payment app on Solana that discovers people near you over AWDL - the radio protocol behind apple's airdrop. basically two phones find each other over direct radio link. tap a face, enter an amount, send sol.

built an payment app on Solana that discovers people near you over AWDL - the radio protocol behind apple's airdrop. basically two phones find each other over direct radio link. tap a face, enter an amount, send sol.

Ash

18,901 views • 4 months ago

Cloud GPU training is a scam. A single M4 MacBook does 2.9 TFLOPS. Seven friends with MacBooks match an NVIDIA A100. Alexander Hayes just open-sourced a tool that makes this work over Wi-Fi. It's called AirTrain. Here's how it works: Traditional distributed training (DDP) syncs gradients after every single step. For a 124M parameter model, that's ~500MB exchanged per step. You need 50 GB/s of sustained bandwidth. Impossible over Wi-Fi. AirTrain uses the DiLoCo algorithm. Each Mac trains independently for 500 steps, then syncs only the difference. One sync per 500 steps instead of one per step. 500x less network communication. Wi-Fi actually works. The entire sync takes ~2 seconds. Here's what makes it wild: → Zero-config discovery. Devices find each other automatically via mDNS/Bonjour. Same protocol as AirDrop. → Fault tolerant. Nodes can join and leave mid-training without killing the run. → Checkpoint relay. Train for a few hours, export a checkpoint, hand it off to someone else to continue. Like a relay race for ML training. → Built on Apple's MLX framework. Native to M1/M2/M3/M4/M5 unified memory. No host-to-device copy overhead. → Local dashboard. Real-time loss curves, peer monitoring, throughput metrics in your browser. Here's the wildest part: An M4 Max with 128GB unified memory can train a 70B parameter model without offloading. An NVIDIA RTX 4090 has 24GB VRAM. Apple Silicon gets ~245-460 GFLOPS per watt. Training on MacBooks costs almost nothing in electricity compared to cloud GPUs. And there are hundreds of millions of Apple Silicon Macs in the world. The math: Traditional DDP: 1 sync per step = 50 GB/s required AirTrain (DiLoCo): 1 sync per 500 steps = 0.1 GB/s required Wi-Fi handles 0.1 GB/s. That's it. That's the breakthrough. They even built a community platform at with live session browsing, checkpoint sharing, and a contributor leaderboard. Training a 124M parameter GPT-2? Instead of renting cloud GPUs at $3/hr, pool three MacBooks in a coffee shop and train for free. MIT licensed. Built in Python. 1 contributor. Early stage but the idea is insane. 100% Open Source. (Link in the comments)

Cloud GPU training is a scam. A single M4 MacBook does 2.9 TFLOPS. Seven friends with MacBooks match an NVIDIA A100. Alexander Hayes just open-sourced a tool that makes this work over Wi-Fi. It's called AirTrain. Here's how it works: Traditional distributed training (DDP) syncs gradients after every single step. For a 124M parameter model, that's ~500MB exchanged per step. You need 50 GB/s of sustained bandwidth. Impossible over Wi-Fi. AirTrain uses the DiLoCo algorithm. Each Mac trains independently for 500 steps, then syncs only the difference. One sync per 500 steps instead of one per step. 500x less network communication. Wi-Fi actually works. The entire sync takes ~2 seconds. Here's what makes it wild: → Zero-config discovery. Devices find each other automatically via mDNS/Bonjour. Same protocol as AirDrop. → Fault tolerant. Nodes can join and leave mid-training without killing the run. → Checkpoint relay. Train for a few hours, export a checkpoint, hand it off to someone else to continue. Like a relay race for ML training. → Built on Apple's MLX framework. Native to M1/M2/M3/M4/M5 unified memory. No host-to-device copy overhead. → Local dashboard. Real-time loss curves, peer monitoring, throughput metrics in your browser. Here's the wildest part: An M4 Max with 128GB unified memory can train a 70B parameter model without offloading. An NVIDIA RTX 4090 has 24GB VRAM. Apple Silicon gets ~245-460 GFLOPS per watt. Training on MacBooks costs almost nothing in electricity compared to cloud GPUs. And there are hundreds of millions of Apple Silicon Macs in the world. The math: Traditional DDP: 1 sync per step = 50 GB/s required AirTrain (DiLoCo): 1 sync per 500 steps = 0.1 GB/s required Wi-Fi handles 0.1 GB/s. That's it. That's the breakthrough. They even built a community platform at with live session browsing, checkpoint sharing, and a contributor leaderboard. Training a 124M parameter GPT-2? Instead of renting cloud GPUs at $3/hr, pool three MacBooks in a coffee shop and train for free. MIT licensed. Built in Python. 1 contributor. Early stage but the idea is insane. 100% Open Source. (Link in the comments)

Guri Singh

160,201 views • 3 months ago

Everyone wrote Apple off as the AI loser, but one hardware spec might flip that story upside down (Save this). @jason called Apple a screaming buy on the back of a single chip detail. The rumored M7 Ultra, expected around 2028, is designed to support up to 1.5TB of unified memory, enough to run frontier class trillion parameter AI models locally, with no cloud required. The Street's bear case on Apple is straightforward. Apple has no frontier model of its own, Siri has stumbled for years and the company effectively rents OpenAI's models for its hardest queries. That narrative treats Apple as the one Magnificent Seven name that missed the AI wave entirely but the bull case flips that framing on its head. If frontier AI models keep shrinking and getting cheaper to run, Apple doesn't need the smartest model in the world, it just needs to own the device that model runs on. And unified memory is the mechanism that makes this possible. Unlike traditional systems where the CPU and GPU each need separate memory, Apple's architecture lets the CPU, GPU and Neural Engine draw from one shared pool. A fully specced M7 Ultra could theoretically run something on the scale of a 1.2 trillion parameter model locally and that capability plugs directly into the one advantage Apple has spent over a decade building: privacy. Apple has already shipped Private Cloud Compute, a system designed so even Apple can't access user data processed off device. Apple doubled down on this at WWDC 2026, framing on device privacy as non-negotiable while rivals default to the cloud. If the best AI models get small enough to run on Apple silicon, the moat stops being the model and becomes the hardware it has to sit on. Milk Road Pro remains bullish on Apple and it remains as one of our core positions, if you want the full thesis + our full AI trades, come join us using the link below for just a $1.

Everyone wrote Apple off as the AI loser, but one hardware spec might flip that story upside down (Save this). @jason called Apple a screaming buy on the back of a single chip detail. The rumored M7 Ultra, expected around 2028, is designed to support up to 1.5TB of unified memory, enough to run frontier class trillion parameter AI models locally, with no cloud required. The Street's bear case on Apple is straightforward. Apple has no frontier model of its own, Siri has stumbled for years and the company effectively rents OpenAI's models for its hardest queries. That narrative treats Apple as the one Magnificent Seven name that missed the AI wave entirely but the bull case flips that framing on its head. If frontier AI models keep shrinking and getting cheaper to run, Apple doesn't need the smartest model in the world, it just needs to own the device that model runs on. And unified memory is the mechanism that makes this possible. Unlike traditional systems where the CPU and GPU each need separate memory, Apple's architecture lets the CPU, GPU and Neural Engine draw from one shared pool. A fully specced M7 Ultra could theoretically run something on the scale of a 1.2 trillion parameter model locally and that capability plugs directly into the one advantage Apple has spent over a decade building: privacy. Apple has already shipped Private Cloud Compute, a system designed so even Apple can't access user data processed off device. Apple doubled down on this at WWDC 2026, framing on device privacy as non-negotiable while rivals default to the cloud. If the best AI models get small enough to run on Apple silicon, the moat stops being the model and becomes the hardware it has to sit on. Milk Road Pro remains bullish on Apple and it remains as one of our core positions, if you want the full thesis + our full AI trades, come join us using the link below for just a $1.

Milk Road AI

37,388 views • 22 days ago

I took Grok 4 for a spin this weekend to build this game prototype. I used SuperGrok Chat to generate the initial game prototype and then brought it over to Cursor to continue coding with Grok 4 MAX. Grok 4 in Cursor is like a no-nonsense agent. Doesn't speak much, but delivers the goods. There were moments where I was rate-limited or stuck on a bug or two that I had to get other models to help, but otherwise it's a fast, reliable model to work with. This makes me incredibly excited for Grok Code to launch in August!

I took Grok 4 for a spin this weekend to build this game prototype. I used SuperGrok Chat to generate the initial game prototype and then brought it over to Cursor to continue coding with Grok 4 MAX. Grok 4 in Cursor is like a no-nonsense agent. Doesn't speak much, but delivers the goods. There were moments where I was rate-limited or stuck on a bug or two that I had to get other models to help, but otherwise it's a fast, reliable model to work with. This makes me incredibly excited for Grok Code to launch in August!

Danny Limanseta

20,484,971 views • 1 year ago

What is Apple doing in the AI race? Ever since ChatGPT came out in 2022, every tech company realized that generative AI is the next big thing. So, all these companies dropped everything else and started focusing on it first. Google launches Bard and does a bunch of stuff. Microsoft teams up with OpenAI and rolls out a pilot. Adobe launches Firefly. Elon Musk starts his new company, XI. Meta launches the Llama model. Tons of other AI startups pop up, and investors are throwing money at AI like crazy Apple's AI strategy is fascinating because it's playing a completely different game than Google, Microsoft, and OpenAI. While everyone else rushed to build the most powerful language models, Apple took a fundamentally different approach that aligns with their core business model and strengths Apple Intelligence is comprised of multiple highly capable generative models that are specialized for users' everyday tasks, but unlike competitors, Apple isn't trying to win the raw AI power race. Instead, they're leveraging what they've always done best, creating seamless, integrated experiences The key insight you mentioned about revenue models is crucial. While Microsoft makes 48% from cloud services and Google relies heavily on cloud and subscriptions, Apple's business is 80% hardware driven. This means they don't need to compete on cloud AI services they can focus on making AI work better on the devices people already own Apple's four step strategy you outlined is spot on, The "Invisible Model" approach is brilliant because most users don't want to think about which AI model to use. Tim Cook doubled down on Apple's AI strategy, insisting that generative AI was never off the table and was always about pursuing it in a thoughtful kind of way, they're making AI feel natural rather than technical Ecosystem Integration remains Apple's superpower. At WWDC 2025, Apple announced what it calls the Foundation Models framework, which will let developers tap into its AI models while offline, this is huge because it means third party apps can now leverage Apple's AI without internet dependency, something Google and Microsoft can't easily replicate across their fragmented hardware ecosystem The Distribution Advantage is where Apple really shines. They have direct control over 2 billion devices with powerful Apple Silicon chips that can run AI models locally. Apple is still pushing App Intents, the same system that makes it simpler for Apple Intelligence and Siri to use apps and get things done, which will enable those complex multi app workflows you described Building Trust through privacy focused messaging is classic Apple. They're positioning themselves as the "safe" AI option while competitors deal with data privacy concerns The real genius is that Apple doesn't need to build the world's best AI model, they just need to build the best AI experience. By partnering with OpenAI for complex tasks while handling simple ones locally, they're creating a hybrid approach that prioritizes user experience over technical bragging rights The upcoming Apple Intelligence features slated for 2025 demonstrate Apple's commitment to integrating advanced AI technologies into its devices, enhancing user experience, and promoting productivity, suggesting they're still in the early phases of a longer term strategy This approach could indeed "wipe out" Android and Windows in the AI era not by building better models, but by making AI feel like a natural extension of the devices people already love. It's classic Apple, arrive late, but redefine the entire category

What is Apple doing in the AI race? Ever since ChatGPT came out in 2022, every tech company realized that generative AI is the next big thing. So, all these companies dropped everything else and started focusing on it first. Google launches Bard and does a bunch of stuff. Microsoft teams up with OpenAI and rolls out a pilot. Adobe launches Firefly. Elon Musk starts his new company, XI. Meta launches the Llama model. Tons of other AI startups pop up, and investors are throwing money at AI like crazy Apple's AI strategy is fascinating because it's playing a completely different game than Google, Microsoft, and OpenAI. While everyone else rushed to build the most powerful language models, Apple took a fundamentally different approach that aligns with their core business model and strengths Apple Intelligence is comprised of multiple highly capable generative models that are specialized for users' everyday tasks, but unlike competitors, Apple isn't trying to win the raw AI power race. Instead, they're leveraging what they've always done best, creating seamless, integrated experiences The key insight you mentioned about revenue models is crucial. While Microsoft makes 48% from cloud services and Google relies heavily on cloud and subscriptions, Apple's business is 80% hardware driven. This means they don't need to compete on cloud AI services they can focus on making AI work better on the devices people already own Apple's four step strategy you outlined is spot on, The "Invisible Model" approach is brilliant because most users don't want to think about which AI model to use. Tim Cook doubled down on Apple's AI strategy, insisting that generative AI was never off the table and was always about pursuing it in a thoughtful kind of way, they're making AI feel natural rather than technical Ecosystem Integration remains Apple's superpower. At WWDC 2025, Apple announced what it calls the Foundation Models framework, which will let developers tap into its AI models while offline, this is huge because it means third party apps can now leverage Apple's AI without internet dependency, something Google and Microsoft can't easily replicate across their fragmented hardware ecosystem The Distribution Advantage is where Apple really shines. They have direct control over 2 billion devices with powerful Apple Silicon chips that can run AI models locally. Apple is still pushing App Intents, the same system that makes it simpler for Apple Intelligence and Siri to use apps and get things done, which will enable those complex multi app workflows you described Building Trust through privacy focused messaging is classic Apple. They're positioning themselves as the "safe" AI option while competitors deal with data privacy concerns The real genius is that Apple doesn't need to build the world's best AI model, they just need to build the best AI experience. By partnering with OpenAI for complex tasks while handling simple ones locally, they're creating a hybrid approach that prioritizes user experience over technical bragging rights The upcoming Apple Intelligence features slated for 2025 demonstrate Apple's commitment to integrating advanced AI technologies into its devices, enhancing user experience, and promoting productivity, suggesting they're still in the early phases of a longer term strategy This approach could indeed "wipe out" Android and Windows in the AI era not by building better models, but by making AI feel like a natural extension of the devices people already love. It's classic Apple, arrive late, but redefine the entire category

D4rsh🦅

13,266 views • 1 year ago

NVIDIA will refund your cloud bill if you let them put a $250,000 supercomputer on your desk for $2,999 The $2,999 DGX Spark puts a 128GB AI supercomputer on your desk the same 70B models you rented in the cloud, but nothing goes over the network and no ToS governs the machine you own $22,000 in annual savings is the big number everyone posts The quiet number is even bigger it’s the contracts you’ve stopped losing due to data residency Owning your own compute is no longer just budget optimization it’s a way to sell to customers who can’t send their data to the cloud

NVIDIA will refund your cloud bill if you let them put a $250,000 supercomputer on your desk for $2,999 The $2,999 DGX Spark puts a 128GB AI supercomputer on your desk the same 70B models you rented in the cloud, but nothing goes over the network and no ToS governs the machine you own $22,000 in annual savings is the big number everyone posts The quiet number is even bigger it’s the contracts you’ve stopped losing due to data residency Owning your own compute is no longer just budget optimization it’s a way to sell to customers who can’t send their data to the cloud

Asteri

143,538 views • 2 months ago

YouTube + Ai = $31,662 • Claude + Manus for the script • Elleven labs for the voice over • Affordable freelancers for the editing I made $31k in the past month of a faceless youtube channel that most people have no idea that I run. Hotel room. Laptop. Wi-Fi. No meetings. No alarm. No boss. Nobody knows what I do. Lowkey rich. f**** it. im giving my entire play-by-play, battle tested, automation dialled process on a 100% live masterclass this week. How to find the 6-figure niches in a process so simple a 5 year old can understand it. Comment "training" and I'll dm ya my registration link to my masterclass (must follow so I can DM you) Let's get you lowkey rich.

YouTube + Ai = $31,662 • Claude + Manus for the script • Elleven labs for the voice over • Affordable freelancers for the editing I made $31k in the past month of a faceless youtube channel that most people have no idea that I run. Hotel room. Laptop. Wi-Fi. No meetings. No alarm. No boss. Nobody knows what I do. Lowkey rich. f**** it. im giving my entire play-by-play, battle tested, automation dialled process on a 100% live masterclass this week. How to find the 6-figure niches in a process so simple a 5 year old can understand it. Comment "training" and I'll dm ya my registration link to my masterclass (must follow so I can DM you) Let's get you lowkey rich.

Devon Canup | YouTube Expert

23,772 views • 4 months ago

We asked (Co-Founder & CEO, webAI) about the advantages of running AI on devices. "There's no free lunch, but there are things cloud players can't do. It's RAM intensive, yes, but by distributing across devices, we can unlock more accuracy and context. But if you're hosting for a million users on Nvidia, you can't do that." "With our stack, everything's downstream only. Think about health data, like you want that to be private. The dream is personalized models for millions that never leave their device."

We asked (Co-Founder & CEO, webAI) about the advantages of running AI on devices. "There's no free lunch, but there are things cloud players can't do. It's RAM intensive, yes, but by distributing across devices, we can unlock more accuracy and context. But if you're hosting for a million users on Nvidia, you can't do that." "With our stack, everything's downstream only. Think about health data, like you want that to be private. The dream is personalized models for millions that never leave their device."

TBPN

10,084 views • 1 year ago

Ben Affleck says actors have no job security because 130K people are in the actors union and only 30,000 work each year "I got real cold and had a couple of movies that didn't work... all of a sudden it can be kind of over for you" "there's no seniority, there's no tenure, there's no retirement. There's no gold watch" "if your show doesn't work or your movie doesn't work and nobody wants to hire you, they're just gonna hire someone else" "there's 130,000 people in SAG and there's only 30,000 people that work every year"

Ben Affleck says actors have no job security because 130K people are in the actors union and only 30,000 work each year "I got real cold and had a couple of movies that didn't work... all of a sudden it can be kind of over for you" "there's no seniority, there's no tenure, there's no retirement. There's no gold watch" "if your show doesn't work or your movie doesn't work and nobody wants to hire you, they're just gonna hire someone else" "there's 130,000 people in SAG and there's only 30,000 people that work every year"

Yonan

5,254,065 views • 2 months ago

My OpenClaw just generated a fully animated AI video with a single Telegram message 🤯 This is V2. And it's a completely different beast. V1 went viral — type a topic, get a fun animated video back. Educational content, silly food characters, entertainment stuff. V2 turns it into an actual ad production system. Two upgrades that change everything: → Upload your own brand mascot or product character → The agent locks that image across every scene via Nano Banana Pro → It writes real direct response scripts — hook, pain points, proof, CTA → Output is built to run as a Facebook ad or native TikTok → Still one message. Still fully automated. No generic educational fluff. No random AI tomato characters. No manual prompting between tools. What you get back: - A 32-second animated video starring YOUR brand character - DR ad script with a real hook, agitation, and CTA - Consistent character across all 4 scenes - Ready to drop into Meta or TikTok as-is Perfect for DTC brands and agencies producing volume creative without a production team. The agent writes the script, generates every frame on-brand, animates with Veo 3, adds voiceover natively, and delivers the final cut to Telegram. I put the V2 skill + full setup guide into a Google Drive folder. Want access for free? > Like this post > Comment "CLAW" And I'll send it over (must be following so I can DM)

Mike Futia

55,615 views • 5 months ago

There are many things we have no power over right now in the world. The situation in Israel is terrifying and I wanted to post some places in the most need if you feel you want to contribute financially. All of the below links are at the link in my bio. -If you want to support the community stormed by terrorists, Kibbutz Nir Or, here is a link that is directly supporting that kibbutz: The direct link is in my bio. -The hospital in Ashkelon that was both hit by a direct rocket and is responsible for treating those in the region closest to the Gaza border, the direct link is in my bio. -The major aid organizations for Israel are -IsraAid: -Hatzalah: -New Israel Fund: -Israel Emergency Fund: These are all distributing resources. -If supporting mental health to a country on its knees is something close to your heart, Trauma4Good assists military and rescue service active members and veterans in getting diagnosed & receiving help for PTSD. -The Base supports lone soldiers, and lone b’not sheirut (young women carrying out national service) They provide practical and emotional support. -There is a civil command center that is organizing and mobilizing supplies throughout the country: meals and shelter to evacuees, equipment to soldiers, etc. Here is the link for a tax-deductible donation (all funds go directly to purchase equipment or pay suppliers) (Please specify that the donation is for the civil command center)

There are many things we have no power over right now in the world. The situation in Israel is terrifying and I wanted to post some places in the most need if you feel you want to contribute financially. All of the below links are at the link in my bio. -If you want to support the community stormed by terrorists, Kibbutz Nir Or, here is a link that is directly supporting that kibbutz: The direct link is in my bio. -The hospital in Ashkelon that was both hit by a direct rocket and is responsible for treating those in the region closest to the Gaza border, the direct link is in my bio. -The major aid organizations for Israel are -IsraAid: -Hatzalah: -New Israel Fund: -Israel Emergency Fund: These are all distributing resources. -If supporting mental health to a country on its knees is something close to your heart, Trauma4Good assists military and rescue service active members and veterans in getting diagnosed & receiving help for PTSD. -The Base supports lone soldiers, and lone b’not sheirut (young women carrying out national service) They provide practical and emotional support. -There is a civil command center that is organizing and mobilizing supplies throughout the country: meals and shelter to evacuees, equipment to soldiers, etc. Here is the link for a tax-deductible donation (all funds go directly to purchase equipment or pay suppliers) (Please specify that the donation is for the civil command center)

Mayim Bialik

819,461 views • 2 years ago

🚨 BOOM: JD Vance says the needs of NO OTHER COUNTRY come before America "My job is to pursue the interests of NO other nation on Earth except for the United States of America!" 🇺🇸 "So when our interests are aligned with Israel? We talk about how to accomplish those SHARED objectives." "When my viewpoint is averse to the Prime Minister of Israel, we have a frank conversation about that. I would frankly describe it as a pleasant but direct conversation. I didn't feel confronted."

🚨 BOOM: JD Vance says the needs of NO OTHER COUNTRY come before America "My job is to pursue the interests of NO other nation on Earth except for the United States of America!" 🇺🇸 "So when our interests are aligned with Israel? We talk about how to accomplish those SHARED objectives." "When my viewpoint is averse to the Prime Minister of Israel, we have a frank conversation about that. I would frankly describe it as a pleasant but direct conversation. I didn't feel confronted."

Eric Daugherty

53,023 views • 5 days ago

A guy just turned standard home Wi-Fi into a radar that sees through walls. Over 62,000 stars on GitHub for a system that tracks humans without using a single camera lens. It doesn’t just know you’re there. Watch what happens when a person lies down on a bed in another room. The AI instantly maps their exact posture, tracks their breathing, and calculates their heartbeat in real-time. No optical surveillance, no privacy leaks - it just analyzes how your body disrupts ordinary radio waves. The whole thing runs on a tiny $5 microchip. Traditional security requires a direct line of sight and invades your personal space. Now, your Wi-Fi router is a camera that sees through solid brick.

A guy just turned standard home Wi-Fi into a radar that sees through walls. Over 62,000 stars on GitHub for a system that tracks humans without using a single camera lens. It doesn’t just know you’re there. Watch what happens when a person lies down on a bed in another room. The AI instantly maps their exact posture, tracks their breathing, and calculates their heartbeat in real-time. No optical surveillance, no privacy leaks - it just analyzes how your body disrupts ordinary radio waves. The whole thing runs on a tiny $5 microchip. Traditional security requires a direct line of sight and invades your personal space. Now, your Wi-Fi router is a camera that sees through solid brick.

pulmencr

1,831,574 views • 21 days ago

The latest MLX has a CUDA back-end! To get started: pip install "mlx[cuda]" With the same codebase you can develop locally, run your model on Apple silicon, or in the cloud on Nvidia GPUs. MLX is designed around Apple silicon - which has a unified memory architecture. It uses the same design with CUDA. So there's no need to move arrays around from CPU memory to GPU memory. Note, this is early days - some operations are missing and performance is still being optimized. But it's already quite fast for Transformer training, text generation, and more! Here's a demo using mlx-lm to generate text with Llama 3 8B (bf16) on an A100:

The latest MLX has a CUDA back-end! To get started: pip install "mlx[cuda]" With the same codebase you can develop locally, run your model on Apple silicon, or in the cloud on Nvidia GPUs. MLX is designed around Apple silicon - which has a unified memory architecture. It uses the same design with CUDA. So there's no need to move arrays around from CPU memory to GPU memory. Note, this is early days - some operations are missing and performance is still being optimized. But it's already quite fast for Transformer training, text generation, and more! Here's a demo using mlx-lm to generate text with Llama 3 8B (bf16) on an A100:

Awni Hannun

42,761 views • 1 year ago

Closing my DM here for a while | my head is going crazy 🤦‍♂️ I’ve verified over 5500 people still have up to 3000 to verify the number keeps increasing after each verification Use the link instead this account verify fast: it’s not be it’s HUMAN PROTOCOL I need to eat, I need to rest but no I must fulfill all approval first 🫳🎤

Closing my DM here for a while | my head is going crazy 🤦‍♂️ I’ve verified over 5500 people still have up to 3000 to verify the number keeps increasing after each verification Use the link instead this account verify fast: it’s not be it’s HUMAN PROTOCOL I need to eat, I need to rest but no I must fulfill all approval first 🫳🎤

Farmercist 👨‍🌾 🦅

81,513 views • 1 year ago

If you are a SaaS Founder and you want to increase your Domain Authority. To get customers from Google and AI on Autopilot. Let me introduce you to Distribb backlink exchange. We don't just swap links randomly like every other network. Which, by the way, doesn't bring you much power to your website. We built a system that: 1. Connects to your Google Search Console 2. Finds your top 5 direct competitors 3. Calculates your backlink gap 4. Matches you with real websites in your niche 5. Writes articles and listicles that link to you (optimized for Google AND AI answers) 6. Distributes links across your homepage, money pages, and blog with natural anchor diversity 7. Publishes them across the network automatically 8. Spreads everything over months so it looks 100% organic 9. Uses A-B-C linking: no direct swaps, no footprints Bonus: The longer you stay, the more links you get. Month 1: 5 backlinks/mo Month 3: 8 backlinks/mo Month 6: 12 backlinks/mo Month 12: 20 backlinks/mo Month 18: 25 backlinks/mo Month 24: 50 backlinks/mo You start at your current DR level. As your authority grows, you unlock links from stronger sites. You focus on building your product. We handle the SEO. There are already hundreds of websites in the network waiting for you. This is not a marketplace where you pay $500 per link. This is not a PBN that gets you penalized. This is a real network of real SaaS websites exchanging real editorial backlinks. Your first month starts at $9 👇

If you are a SaaS Founder and you want to increase your Domain Authority. To get customers from Google and AI on Autopilot. Let me introduce you to Distribb backlink exchange. We don't just swap links randomly like every other network. Which, by the way, doesn't bring you much power to your website. We built a system that: 1. Connects to your Google Search Console 2. Finds your top 5 direct competitors 3. Calculates your backlink gap 4. Matches you with real websites in your niche 5. Writes articles and listicles that link to you (optimized for Google AND AI answers) 6. Distributes links across your homepage, money pages, and blog with natural anchor diversity 7. Publishes them across the network automatically 8. Spreads everything over months so it looks 100% organic 9. Uses A-B-C linking: no direct swaps, no footprints Bonus: The longer you stay, the more links you get. Month 1: 5 backlinks/mo Month 3: 8 backlinks/mo Month 6: 12 backlinks/mo Month 12: 20 backlinks/mo Month 18: 25 backlinks/mo Month 24: 50 backlinks/mo You start at your current DR level. As your authority grows, you unlock links from stronger sites. You focus on building your product. We handle the SEO. There are already hundreds of websites in the network waiting for you. This is not a marketplace where you pay $500 per link. This is not a PBN that gets you penalized. This is a real network of real SaaS websites exchanging real editorial backlinks. Your first month starts at $9 👇

Florian Darroman

26,669 views • 4 months ago

I had a voice chat with founder of Humanity this morning. I want to be very transparent and keep everything up to you, that is why I am publishing this video recording even though he asked me not to. I want to let all VCs and Web3 people know the founder's mentality about their users. TLDR : 1) They reserved some percentage of allocation from Community Incentive (which was supposed to be distribute among users) for CEX listing 2) If you published your referral link publicly and any bad actor joined through it, your allocation was significantly penalized. You were expected to share it only with friends you trust. 3) They distributed the airdrop based on the value of your community. For example, if a project founder got 100 referrals and an airdrop influencer got the same number, the founder’s community was considered more valuable, so the founder received a larger allocation than a regular airdrop influencer. 4) If a bad actor joined through your referral link, or if you joined through a bad actor’s referral link, both of you were penalized or deemed ineligible for the airdrop. 5) A few minutes ago, I checked that they updated my allocation to 20,565 $H, while other users’ allocations increased by only 100 $H.

I had a voice chat with founder of Humanity this morning. I want to be very transparent and keep everything up to you, that is why I am publishing this video recording even though he asked me not to. I want to let all VCs and Web3 people know the founder's mentality about their users. TLDR : 1) They reserved some percentage of allocation from Community Incentive (which was supposed to be distribute among users) for CEX listing 2) If you published your referral link publicly and any bad actor joined through it, your allocation was significantly penalized. You were expected to share it only with friends you trust. 3) They distributed the airdrop based on the value of your community. For example, if a project founder got 100 referrals and an airdrop influencer got the same number, the founder’s community was considered more valuable, so the founder received a larger allocation than a regular airdrop influencer. 4) If a bad actor joined through your referral link, or if you joined through a bad actor’s referral link, both of you were penalized or deemed ineligible for the airdrop. 5) A few minutes ago, I checked that they updated my allocation to 20,565 $H, while other users’ allocations increased by only 100 $H.

Zun

133,500 views • 1 year ago

TODAY IS TAX DAY. HERE'S WHAT WARREN BUFFETT SAID ABOUT PAYING TAXES IN 2024 "Berkshire sent in a check for over $5 billion to the U.S. federal government last year." "If 800 other companies had done the same thing, no other person in the United States would have had to pay a dime of federal taxes. No income taxes. No Social Security taxes. No estate taxes. Nothing." "It doesn't bother me in the least to write that check. And I would really hope, with all that America has done for all of you, it shouldn't bother you that we do it."

TODAY IS TAX DAY. HERE'S WHAT WARREN BUFFETT SAID ABOUT PAYING TAXES IN 2024 "Berkshire sent in a check for over $5 billion to the U.S. federal government last year." "If 800 other companies had done the same thing, no other person in the United States would have had to pay a dime of federal taxes. No income taxes. No Social Security taxes. No estate taxes. Nothing." "It doesn't bother me in the least to write that check. And I would really hope, with all that America has done for all of you, it shouldn't bother you that we do it."

WOLF

421,373 views • 3 months ago

The World Cup + Sports betting affiliate is f*cking gold mine I made $20k with a TikTok account, a VA and Google Omni it takes me 2 hours a day to generate about 20 videos and costs me about $0.30 per video the best part is there's is literally 0 competition because no one knows how to make AI videos that look this good for this cheap All I do is: Get the script from a Claude skill I wrote, edit it a bit and send it to my jeet who then generates me the images based of the script using Pinterest and GPT 2 She then sends me the images, I generate the full AI video using my Omni method and I upload them across my TikTok accounts which funnels them to my domain The crazy thing is literally anyone could come in and run this by themselves and a few hours spare a day I put the whole method into a video for everyone including >How to generate AI UGC that is realistic enough to trick a 18 year old >How to creative direct a TikTok video so it goes viral >How to structure yapper style content to actually convert and go viral >The offer I'm using and my Claude skill I wrote as a bonus RT + Comment "YAPP" and I'll send you the video over (must be following so I can DM)

The World Cup + Sports betting affiliate is f*cking gold mine I made $20k with a TikTok account, a VA and Google Omni it takes me 2 hours a day to generate about 20 videos and costs me about $0.30 per video the best part is there's is literally 0 competition because no one knows how to make AI videos that look this good for this cheap All I do is: Get the script from a Claude skill I wrote, edit it a bit and send it to my jeet who then generates me the images based of the script using Pinterest and GPT 2 She then sends me the images, I generate the full AI video using my Omni method and I upload them across my TikTok accounts which funnels them to my domain The crazy thing is literally anyone could come in and run this by themselves and a few hours spare a day I put the whole method into a video for everyone including >How to generate AI UGC that is realistic enough to trick a 18 year old >How to creative direct a TikTok video so it goes viral >How to structure yapper style content to actually convert and go viral >The offer I'm using and my Claude skill I wrote as a bonus RT + Comment "YAPP" and I'll send you the video over (must be following so I can DM)

Pounds

41,440 views • 1 month ago

217 mph ball speed. 140+ mph clubhead speed. Now teeing it up at Oakmont. Last night on the Golf Channel, Matt Vogt shared how his training with Michael Christman helped him add 15 lbs to his 6’6” frame and over 20 mph to his max ball speed — all while balancing a full-time career as a dentist. This week, he’s bringing that speed to the U.S. Open — at the very course where he caddied growing up. While there is no shortage of useful information related to golf performance on the internet, Matt’s success is proof that there is no substitute for hiring a qualified professional who can help put you on a path towards your goals.

217 mph ball speed. 140+ mph clubhead speed. Now teeing it up at Oakmont. Last night on the Golf Channel, Matt Vogt shared how his training with Michael Christman helped him add 15 lbs to his 6’6” frame and over 20 mph to his max ball speed — all while balancing a full-time career as a dentist. This week, he’s bringing that speed to the U.S. Open — at the very course where he caddied growing up. While there is no shortage of useful information related to golf performance on the internet, Matt’s success is proof that there is no substitute for hiring a qualified professional who can help put you on a path towards your goals.

TPI

13,577 views • 1 year ago