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.

This is from Apple's State of the Union The local model is a 3B parameter SLM that uses adapters trained for each specific feature. Diffusion model does the same thing, adapter for each style. Anything running locally or Apple's Secure Cloud is an Apple model, not OpenAI.

Max Weinbach

302,624 subscribers

2,648,372 views • 2 years ago •via X (Twitter)

Education Science & Technology

Anya Rossi• Live Now

Private livecam show

10 Comments

Pierson Marks2 years ago

If I have an Macbook Pro nearby, could we run larger models "semi-locally" via bluetooth communication and share compute?

Max Weinbach2 years ago

I was hoping for this or even better HomePods with M series chips, but my guess is they’ll need dedicated hardware and networking

Ryan Boodram 🇬🇾🥥🌴2 years ago

@CommunityNotes cite this under all the posts saying Apple didn’t develop their own AI or are just a “ChatGPT wrapper”

bittermelon2 years ago

@elonmusk the local model IS NOT OpenAI

Tux2 years ago

some people with stupid hate on Apple, the presentation was very clear on what would be using ChatGPT.

CorruptedBirdie2 years ago

Didn’t take long for the shills to defend a $3 trillion dollar company that doesn’t even know they exist. 😂😂

Max Weinbach2 years ago

Two things: 1. I’m spreading more information and truth about how it works 2. They know me

Gordus2 years ago

Specific details directly from Apple are here:

Marek2 years ago

@elonmusk

Parker2 years ago

Hey @elonmusk might want to take a 6:31 break and watch this

Related Videos

Apple built a large foundation model and fine-tuned it on multiple tasks. But they are doing something very clever: They load a single model in memory and use different adapters to specialize the model on the fly. I recorded a video to show you how to write the code to do the same thing Apple is doing. I explain everything step by step. Here is what I'll show you in the video: 1. We'll load two datasets 2. Then load a large model 3. Then, we'll fine-tune the model on both datasets I'll use LoRA to fine-tune the model. This process creates two small adapters, each specializing in solving one of the datasets. The base model's original parameters will remain unchanged. From here: 4. We'll generate a list of tasks 5. We'll load the correct adapter to solve each task The large model I'm using needs 346 MB of memory, but I only need to load it once. Each adapter is only 2.7 MB. I only need to load the base model once and pair it with any of the fine-tuned adapters. Minimum memory footprint and I can solve multiple tasks. Hope this helps!

Apple built a large foundation model and fine-tuned it on multiple tasks. But they are doing something very clever: They load a single model in memory and use different adapters to specialize the model on the fly. I recorded a video to show you how to write the code to do the same thing Apple is doing. I explain everything step by step. Here is what I'll show you in the video: 1. We'll load two datasets 2. Then load a large model 3. Then, we'll fine-tune the model on both datasets I'll use LoRA to fine-tune the model. This process creates two small adapters, each specializing in solving one of the datasets. The base model's original parameters will remain unchanged. From here: 4. We'll generate a list of tasks 5. We'll load the correct adapter to solve each task The large model I'm using needs 346 MB of memory, but I only need to load it once. Each adapter is only 2.7 MB. I only need to load the base model once and pair it with any of the fine-tuned adapters. Minimum memory footprint and I can solve multiple tasks. Hope this helps!

Santiago

84,747 views • 2 years ago

We have isolated the vocals of over 28,000 songs: Not for any music AI model at all. Not to clone these vocals or the singer’s style. Not anything musical, although I have made some interesting discoveries. It is a new AI model that is being trained on a new paradigm.

We have isolated the vocals of over 28,000 songs: Not for any music AI model at all. Not to clone these vocals or the singer’s style. Not anything musical, although I have made some interesting discoveries. It is a new AI model that is being trained on a new paradigm.

Brian Roemmele

56,584 views • 4 months ago

"This whole story and narrative that Google has already won is going to be harder to tell when the best model is actually an OpenAI model," D.A. Davidson's Gil Luria says. He also says that OpenAI will launch that "state-of-the-art" model "over the next few days, weeks."

"This whole story and narrative that Google has already won is going to be harder to tell when the best model is actually an OpenAI model," D.A. Davidson's Gil Luria says. He also says that OpenAI will launch that "state-of-the-art" model "over the next few days, weeks."

Yahoo Finance

12,410 views • 6 months ago

Been experimenting with Apple's new Core AI framework over the past few days Built a mobile app that runs local AI models directly on device, exploring Core AI model deployment (.aimodel), model specialization, caching, and inference across Apple Silicon Also spent some time integrating with Apple's Foundation Models framework and understanding how custom models can coexist with Apple's on-device AI stack Still early, but the combination of Core AI, Foundation Models, and the Apple Neural Engine feels like a pretty compelling foundation for local first AI applications Still in alpha and working towards a more stable version, but it's been a fun way to explore where Apple's on device AI ecosystem is headed !!

Been experimenting with Apple's new Core AI framework over the past few days Built a mobile app that runs local AI models directly on device, exploring Core AI model deployment (.aimodel), model specialization, caching, and inference across Apple Silicon Also spent some time integrating with Apple's Foundation Models framework and understanding how custom models can coexist with Apple's on-device AI stack Still early, but the combination of Core AI, Foundation Models, and the Apple Neural Engine feels like a pretty compelling foundation for local first AI applications Still in alpha and working towards a more stable version, but it's been a fun way to explore where Apple's on device AI ecosystem is headed !!

Sagar

12,828 views • 1 month ago

DeepSeek V4 Flash + LFM2.5-VL-3B as the vision auxiliary in Hermes Agent is a deadly combo. DeepSeek handles the agentic work. LFM2.5-VL-3B gives it the eyes. I told it to use only its local vision model to browse. It read Hacker News, picked the top story, opened the eclipse map, found an Iceland webcam marker and opened the live feed. Every step seen through a 3B model running locally. No cloud vision API.

DeepSeek V4 Flash + LFM2.5-VL-3B as the vision auxiliary in Hermes Agent is a deadly combo. DeepSeek handles the agentic work. LFM2.5-VL-3B gives it the eyes. I told it to use only its local vision model to browse. It read Hacker News, picked the top story, opened the eclipse map, found an Iceland webcam marker and opened the live feed. Every step seen through a 3B model running locally. No cloud vision API.

Noctus

65,781 views • 3 days ago

Cursor now natively supports git worktrees. You can run multiple agents in parallel, each with their own isolated copy of the codebase. This is great for comparing models or running the same model multiple times and picking the best result.

Cursor now natively supports git worktrees. You can run multiple agents in parallel, each with their own isolated copy of the codebase. This is great for comparing models or running the same model multiple times and picking the best result.

Lee Robinson

157,666 views • 9 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 • 27 days ago

Sam Altman: "if we can make an AI system that is materially better than all of OpenAI at doing AI research, that does feel like an important discontinuity... the model is going to get so good so fast... plan for the model to get rapidly smarter"

Sam Altman: "if we can make an AI system that is materially better than all of OpenAI at doing AI research, that does feel like an important discontinuity... the model is going to get so good so fast... plan for the model to get rapidly smarter"

Tsarathustra

231,733 views • 1 year ago

This may look like a game of Counterstrike running slowly, but it actually extraordinary. The entire game is created, frame-by-frame, on my *home computer* by an AI diffusion model in response to my actions. There is no game engine, just a "world model" trained on Counterstrike

This may look like a game of Counterstrike running slowly, but it actually extraordinary. The entire game is created, frame-by-frame, on my home computer by an AI diffusion model in response to my actions. There is no game engine, just a "world model" trained on Counterstrike

Ethan Mollick

376,399 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

THIS $1,300 CARD HAS THE SAME 32GB OF VRAM AS A $2,000+ RTX 5090 - FOR LOCAL AI, THAT'S THE SPEC THAT DECIDES EVERYTHING. that clip is the gigabyte radeon ai pro r9700 ai top, 32gb of vram, built for local ai. the pitch is simple: the same memory capacity as an rtx 5090, for roughly a third of what a 5090 actually sells for. here's why that matters, and where it doesn't. for ai, vram is the gate. a model either fits in memory or it doesn't. 32gb loads models a 16gb card can't touch, full stop. so for running local llms and image models, this card does the one thing that counts, cheap. honest line: this is not a 5090 for gaming. the 5090 is far faster at raster and ray tracing. they only match on how much vram they carry. if you want frames, buy the 5090. if you want to fit big models locally, you just found the same 32gb for a third of the money. no $2,000+ price tag, no cloud gpu rental, no model too big for 16gb. the uncomfortable part isn't that amd is cheaper. it's that for local ai you were about to overpay for gaming power you'd never use. follow RetroChainer for more on paying for the spec that matters, and bookmark this before your next gpu.

THIS $1,300 CARD HAS THE SAME 32GB OF VRAM AS A $2,000+ RTX 5090 - FOR LOCAL AI, THAT'S THE SPEC THAT DECIDES EVERYTHING. that clip is the gigabyte radeon ai pro r9700 ai top, 32gb of vram, built for local ai. the pitch is simple: the same memory capacity as an rtx 5090, for roughly a third of what a 5090 actually sells for. here's why that matters, and where it doesn't. for ai, vram is the gate. a model either fits in memory or it doesn't. 32gb loads models a 16gb card can't touch, full stop. so for running local llms and image models, this card does the one thing that counts, cheap. honest line: this is not a 5090 for gaming. the 5090 is far faster at raster and ray tracing. they only match on how much vram they carry. if you want frames, buy the 5090. if you want to fit big models locally, you just found the same 32gb for a third of the money. no $2,000+ price tag, no cloud gpu rental, no model too big for 16gb. the uncomfortable part isn't that amd is cheaper. it's that for local ai you were about to overpay for gaming power you'd never use. follow RetroChainer for more on paying for the spec that matters, and bookmark this before your next gpu.

RetroChainer

38,844 views • 7 days ago

Google presents Still-Moving Customized Video Generation without Customized Video Data Customizing text-to-image (T2I) models has seen tremendous progress recently, particularly in areas such as personalization, stylization, and conditional generation. However, expanding this progress to video generation is still in its infancy, primarily due to the lack of customized video data. In this work, we introduce Still-Moving, a novel generic framework for customizing a text-to-video (T2V) model, without requiring any customized video data. The framework applies to the prominent T2V design where the video model is built over a text-to-image (T2I) model (e.g., via inflation). We assume access to a customized version of the T2I model, trained only on still image data (e.g., using DreamBooth or StyleDrop). Naively plugging in the weights of the customized T2I model into the T2V model often leads to significant artifacts or insufficient adherence to the customization data. To overcome this issue, we train lightweight Spatial Adapters that adjust the features produced by the injected T2I layers. Importantly, our adapters are trained on "frozen videos" (i.e., repeated images), constructed from image samples generated by the customized T2I model. This training is facilitated by a novel Motion Adapter module, which allows us to train on such static videos while preserving the motion prior of the video model. At test time, we remove the Motion Adapter modules and leave in only the trained Spatial Adapters. This restores the motion prior of the T2V model while adhering to the spatial prior of the customized T2I model. We demonstrate the effectiveness of our approach on diverse tasks including personalized, stylized, and conditional generation. In all evaluated scenarios, our method seamlessly integrates the spatial prior of the customized T2I model with a motion prior supplied by the T2V model.

Google presents Still-Moving Customized Video Generation without Customized Video Data Customizing text-to-image (T2I) models has seen tremendous progress recently, particularly in areas such as personalization, stylization, and conditional generation. However, expanding this progress to video generation is still in its infancy, primarily due to the lack of customized video data. In this work, we introduce Still-Moving, a novel generic framework for customizing a text-to-video (T2V) model, without requiring any customized video data. The framework applies to the prominent T2V design where the video model is built over a text-to-image (T2I) model (e.g., via inflation). We assume access to a customized version of the T2I model, trained only on still image data (e.g., using DreamBooth or StyleDrop). Naively plugging in the weights of the customized T2I model into the T2V model often leads to significant artifacts or insufficient adherence to the customization data. To overcome this issue, we train lightweight Spatial Adapters that adjust the features produced by the injected T2I layers. Importantly, our adapters are trained on "frozen videos" (i.e., repeated images), constructed from image samples generated by the customized T2I model. This training is facilitated by a novel Motion Adapter module, which allows us to train on such static videos while preserving the motion prior of the video model. At test time, we remove the Motion Adapter modules and leave in only the trained Spatial Adapters. This restores the motion prior of the T2V model while adhering to the spatial prior of the customized T2I model. We demonstrate the effectiveness of our approach on diverse tasks including personalized, stylized, and conditional generation. In all evaluated scenarios, our method seamlessly integrates the spatial prior of the customized T2I model with a motion prior supplied by the T2V model.

AK

40,485 views • 2 years ago

SITUATION EXPLAINED: Why is the Bridgewater Thinking Machines paper a blueprint for every enterprise that has ever paid frontier model prices for generic outputs? 47fucb4r8curb4fc8f8r4bfic8r: "What Thinking Machines and Bridgewater did together is a blueprint for how a wide variety of stakeholders within the market can benefit tremendously from using LLMs and integrating them into a larger enterprise system." "They got a pretty old model. A Qwen3 model. Two hundred and thirty-five billion parameters. Not a new model. Not a very sophisticated or impressive state-of-the-art model. And they essentially fine-tuned it on Bridgewater's own proprietary data." "A financial analyst gets a ton of financial documents and has to filter which are worth looking at. Then evaluate which news pieces are relevant. They have a lot of tacit knowledge from doing the job on where to look. What they did was inject a lot of this tacit knowledge into the model through fine-tuning." "Stop obsessing over state-of-the-art models that are the most intelligent possible. Focus on targeting a very specific LLM specifically trained for a specific function to do that function better than the alternatives on the market." "This old non-state-of-the-art model got 84.7% accuracy versus 78% for all the state-of-the-art models. A massive improvement. And the kicker is they got a better performing model that is 14 times cheaper to use."

SITUATION EXPLAINED: Why is the Bridgewater Thinking Machines paper a blueprint for every enterprise that has ever paid frontier model prices for generic outputs? 47fucb4r8curb4fc8f8r4bfic8r: "What Thinking Machines and Bridgewater did together is a blueprint for how a wide variety of stakeholders within the market can benefit tremendously from using LLMs and integrating them into a larger enterprise system." "They got a pretty old model. A Qwen3 model. Two hundred and thirty-five billion parameters. Not a new model. Not a very sophisticated or impressive state-of-the-art model. And they essentially fine-tuned it on Bridgewater's own proprietary data." "A financial analyst gets a ton of financial documents and has to filter which are worth looking at. Then evaluate which news pieces are relevant. They have a lot of tacit knowledge from doing the job on where to look. What they did was inject a lot of this tacit knowledge into the model through fine-tuning." "Stop obsessing over state-of-the-art models that are the most intelligent possible. Focus on targeting a very specific LLM specifically trained for a specific function to do that function better than the alternatives on the market." "This old non-state-of-the-art model got 84.7% accuracy versus 78% for all the state-of-the-art models. A massive improvement. And the kicker is they got a better performing model that is 14 times cheaper to use."

MTS

93,185 views • 1 month ago

Alex Finn reveals that you need to spend $20,000 to get the equivalent of Opus 4.6 on a local model "The advantages are that you aren’t paying for tokens, all you have to pay for is electricity.” "The fact you're not paying for tokens means you can run your model 24/7 365, you can't do that with Opus" “If you run it locally it’s a completely private system. Nothing you say is private if you're not on a local model”

Alex Finn reveals that you need to spend $20,000 to get the equivalent of Opus 4.6 on a local model "The advantages are that you aren’t paying for tokens, all you have to pay for is electricity.” "The fact you're not paying for tokens means you can run your model 24/7 365, you can't do that with Opus" “If you run it locally it’s a completely private system. Nothing you say is private if you're not on a local model”

Mikli

85,206 views • 6 months ago

$A crypto project actually trained a 72B parameter AI model from scratch using decentralized GPU compute. Not fine-tuned, not a wrapper: trained from zero. The model benchmarks competitively against Meta's LLaMA 3 on reasoning tasks, and the entire training run cost a fraction of what centralized labs spend. If decentralized compute can produce frontier-class models, the moat around OpenAI and Anthropic is thinner than people think.$

A crypto project actually trained a 72B parameter AI model from scratch using decentralized GPU compute. Not fine-tuned, not a wrapper: trained from zero. The model benchmarks competitively against Meta's LLaMA 3 on reasoning tasks, and the entire training run cost a fraction of what centralized labs spend. If decentralized compute can produce frontier-class models, the moat around OpenAI and Anthropic is thinner than people think.

VirtualBacon

40,647 views • 4 months ago

Google just released Muse: a text to image model achieving state of the art performance. It is significantly more efficient than diffusion models. Muse also allows for inpainting, outpainting, and mask-free editing without finetuning or inverting the model.

Google just released Muse: a text to image model achieving state of the art performance. It is significantly more efficient than diffusion models. Muse also allows for inpainting, outpainting, and mask-free editing without finetuning or inverting the model.

bleedingedge.ai

183,750 views • 3 years ago

The entire AI industry is racing to build the smartest model. Satya Nadella just admitted that is not where the money is. The model is not the product. The harness is. That is the exact line. And it changes what Microsoft is actually competing on. OpenAI, Anthropic, Google, xAI, Meta every frontier lab is pouring hundreds of billions into training compute, chasing the next capability jump. Each betting that raw model intelligence is the moat. Microsoft is doing the opposite. It is building the harness the orchestration layer that sits above the model, connecting it to tools, data, permissions, sub-agents, and enterprise workflows. And it is letting OpenAI, Anthropic, and MAI compete to plug into it. "You need the model. But the model is not the product. The harness is." So do the math on what a harness actually does. A raw model dropped into an enterprise answers questions. That is a chatbot. A harness turns that same model into an agent that reads the SharePoint, edits the ERP entry, pulls the GitHub PR, updates Salesforce, and files the Excel report with the right permissions, the right audit trail, and the right sub-agent for each sub-task. The model provides the intelligence. The harness converts intelligence into work. Now here's where it gets interesting. "Even the best model in the world will feel broken without a great harness. And an okay model with a great harness can feel like magic." If that is true, the enterprise buyer is not buying model quality. The enterprise buyer is buying the harness. Which means model quality becomes a commodity input over time, and harness quality becomes the sustainable moat. Compare that to the strategy the entire frontier lab industry is executing. Everyone else is chasing the numerator raw intelligence. Almost nobody at scale is racing to build the denominator the orchestration layer that determines whether that intelligence can actually be deployed profitably inside a real company. The frontier model race has a 10 to 20 percent chance of producing a single dominant winner. Nadella just told the industry he does not need to be that winner. If OpenAI wins, Microsoft wins. If Anthropic wins, Microsoft wins. If MAI wins, Microsoft wins. If someone Microsoft has never heard of trains a better model in 2027, Microsoft still wins. Because the compute they train on, the harness they get plugged into, the enterprise contracts they get delivered through, and the products they sit inside are all Microsoft. He is not building the best AI model. He is building the layer that the best AI model has to run on to make anyone money. I wonder which position looks more valuable in ten years.

The entire AI industry is racing to build the smartest model. Satya Nadella just admitted that is not where the money is. The model is not the product. The harness is. That is the exact line. And it changes what Microsoft is actually competing on. OpenAI, Anthropic, Google, xAI, Meta every frontier lab is pouring hundreds of billions into training compute, chasing the next capability jump. Each betting that raw model intelligence is the moat. Microsoft is doing the opposite. It is building the harness the orchestration layer that sits above the model, connecting it to tools, data, permissions, sub-agents, and enterprise workflows. And it is letting OpenAI, Anthropic, and MAI compete to plug into it. "You need the model. But the model is not the product. The harness is." So do the math on what a harness actually does. A raw model dropped into an enterprise answers questions. That is a chatbot. A harness turns that same model into an agent that reads the SharePoint, edits the ERP entry, pulls the GitHub PR, updates Salesforce, and files the Excel report with the right permissions, the right audit trail, and the right sub-agent for each sub-task. The model provides the intelligence. The harness converts intelligence into work. Now here's where it gets interesting. "Even the best model in the world will feel broken without a great harness. And an okay model with a great harness can feel like magic." If that is true, the enterprise buyer is not buying model quality. The enterprise buyer is buying the harness. Which means model quality becomes a commodity input over time, and harness quality becomes the sustainable moat. Compare that to the strategy the entire frontier lab industry is executing. Everyone else is chasing the numerator raw intelligence. Almost nobody at scale is racing to build the denominator the orchestration layer that determines whether that intelligence can actually be deployed profitably inside a real company. The frontier model race has a 10 to 20 percent chance of producing a single dominant winner. Nadella just told the industry he does not need to be that winner. If OpenAI wins, Microsoft wins. If Anthropic wins, Microsoft wins. If MAI wins, Microsoft wins. If someone Microsoft has never heard of trains a better model in 2027, Microsoft still wins. Because the compute they train on, the harness they get plugged into, the enterprise contracts they get delivered through, and the products they sit inside are all Microsoft. He is not building the best AI model. He is building the layer that the best AI model has to run on to make anyone money. I wonder which position looks more valuable in ten years.

Vikram M

21,463 views • 1 month ago

"This is the first time that OpenAI has released a new model that was not decisively the best." @jason and Gavin Baker discuss OpenAI's GPT-5 release: Jason: " Sam did this tweet, it kind of built a lot of expectation that this would be otherworldly. It wasn't." "Are we hitting either the trough of despair or maybe diminishing returns in the LLM space, where maybe it's feeling incremental, not groundbreaking, when we release these iterations?" Gavin: "Maybe for OpenAI." "This is the first time that OpenAI has released a new model that was not decisively the best." " The one thing about OpenAI is they're very good at product. Just really, really good." "And I just always go back to the statement from Eric Vishria, you know, it might be that if OpenAI never releases another model, they're still a really valuable company."

"This is the first time that OpenAI has released a new model that was not decisively the best." @jason and Gavin Baker discuss OpenAI's GPT-5 release: Jason: " Sam did this tweet, it kind of built a lot of expectation that this would be otherworldly. It wasn't." "Are we hitting either the trough of despair or maybe diminishing returns in the LLM space, where maybe it's feeling incremental, not groundbreaking, when we release these iterations?" Gavin: "Maybe for OpenAI." "This is the first time that OpenAI has released a new model that was not decisively the best." " The one thing about OpenAI is they're very good at product. Just really, really good." "And I just always go back to the statement from Eric Vishria, you know, it might be that if OpenAI never releases another model, they're still a really valuable company."

The All-In Podcast

49,179 views • 1 year ago