Loading video...

Video Failed to Load

Go Home

🔥HOLY SMOKES! $TAO holders! 🚀 SUBNET 19 (VISION) ON BITTENSOR IS ABSOLUTELY CRUSHING IT! In my 5+ years covering crypto and AI, this is one of the most impressive implementations I've seen. The combination of scale, performance, and decentralization is absolutely next level! 🚀 @namoray_dev @Corcel_X 💨 INSANE Speed...

bittingthembits's profile picture

Andy ττ

10,979 subscribers

11,616 views • 1 year ago •via X (Twitter)

Science & TechnologyNews & PoliticsFinance
LIVE

Anya Rossi• Live Now

Private livecam show

3 Comments

TOP_G_CRYPTO's profile picture
TOP_G_CRYPTO1 year ago

LFG TAO!

David's profile picture
David1 year ago

LFG TAO

Vibetech1's profile picture
Vibetech11 year ago

This is absolutely mind blowing information. Thank you for sharing!

Related Videos

DeepSeek R1 671B is ridiculously good, but running at 158 tokens per second is something out of this world. A model *this good* running *this fast* is 3 years ahead of this time. The secret: DeepSeek in this short video isn't running on GPUs. It's running on custom-made chips! This is running on SambaNova and their SN40L chips, which are designed and optimized for running AI workflows. SambaNova is sponsoring my work. We designed GPUs to run games and later realized we could use them for AI. This is very different from creating a chip specifically for AI from the ground up. Just think about this for a second: One single SNL40 chip can simultaneously hold 100+ models (trillions of parameters) in memory! An agentic workflow that uses multiple models simultaneously won't need to swap models from memory even once! By the way, the Llama family of models *screams* on the SambaNova Cloud. Here are some numbers from my latest tests: • Llama 3.3 70B: 367.32 t/s • Llama 3.2 1B: 2381.82 t/s • Llama 3.2 3B: 1335.71 t/s • Llama 3.1 8B: 929.58 t/s • Llama 3.1 70B: 369.16 t/s • Llama 3.1 405B: 103.37 t/s You can access all of these models for free!
1:34

DeepSeek R1 671B is ridiculously good, but running at 158 tokens per second is something out of this world. A model *this good* running *this fast* is 3 years ahead of this time. The secret: DeepSeek in this short video isn't running on GPUs. It's running on custom-made chips! This is running on SambaNova and their SN40L chips, which are designed and optimized for running AI workflows. SambaNova is sponsoring my work. We designed GPUs to run games and later realized we could use them for AI. This is very different from creating a chip specifically for AI from the ground up. Just think about this for a second: One single SNL40 chip can simultaneously hold 100+ models (trillions of parameters) in memory! An agentic workflow that uses multiple models simultaneously won't need to swap models from memory even once! By the way, the Llama family of models *screams* on the SambaNova Cloud. Here are some numbers from my latest tests: • Llama 3.3 70B: 367.32 t/s • Llama 3.2 1B: 2381.82 t/s • Llama 3.2 3B: 1335.71 t/s • Llama 3.1 8B: 929.58 t/s • Llama 3.1 70B: 369.16 t/s • Llama 3.1 405B: 103.37 t/s You can access all of these models for free!

Santiago

131,019 views • 1 year ago

Introducing "Building with Llama 4." This short course is created with Meta AI at Meta, and taught by Amit Sangani, Director of Partner Engineering for Meta’s AI team. Meta’s new Llama 4 has added three new models and introduced the Mixture-of-Experts (MoE) architecture to its family of open-weight models, making them more efficient to serve. In this course, you’ll work with two of the three new models introduced in Llama 4. First is Maverick, a 400B parameter model, with 128 experts and 17B active parameters. Second is Scout, a 109B parameter model with 16 experts and 17B active parameters. Maverick and Scout support long context windows of up to a million tokens and 10M tokens, respectively. The latter is enough to support directly inputting even fairly large GitHub repos for analysis! In hands-on lessons, you’ll build apps using Llama 4’s new multimodal capabilities including reasoning across multiple images and image grounding, in which you can identify elements in images. You’ll also use the official Llama API, work with Llama 4’s long-context abilities, and learn about Llama’s newest open-source tools: its prompt optimization tool that automatically improves system prompts and synthetic data kit that generates high-quality datasets for fine-tuning. If you need an open model, Llama is a great option, and the Llama 4 family is an important part of any GenAI developer's toolkit. Through this course, you’ll learn to call Llama 4 via API, use its optimization tools, and build features that span text, images, and large context. Please sign up here:
3:22

Introducing "Building with Llama 4." This short course is created with Meta AI at Meta, and taught by Amit Sangani, Director of Partner Engineering for Meta’s AI team. Meta’s new Llama 4 has added three new models and introduced the Mixture-of-Experts (MoE) architecture to its family of open-weight models, making them more efficient to serve. In this course, you’ll work with two of the three new models introduced in Llama 4. First is Maverick, a 400B parameter model, with 128 experts and 17B active parameters. Second is Scout, a 109B parameter model with 16 experts and 17B active parameters. Maverick and Scout support long context windows of up to a million tokens and 10M tokens, respectively. The latter is enough to support directly inputting even fairly large GitHub repos for analysis! In hands-on lessons, you’ll build apps using Llama 4’s new multimodal capabilities including reasoning across multiple images and image grounding, in which you can identify elements in images. You’ll also use the official Llama API, work with Llama 4’s long-context abilities, and learn about Llama’s newest open-source tools: its prompt optimization tool that automatically improves system prompts and synthetic data kit that generates high-quality datasets for fine-tuning. If you need an open model, Llama is a great option, and the Llama 4 family is an important part of any GenAI developer's toolkit. Through this course, you’ll learn to call Llama 4 via API, use its optimization tools, and build features that span text, images, and large context. Please sign up here:

Andrew Ng

67,846 views • 1 year ago

🔥 We’re excited to announce that the Finney Testnet is thriving! In record time, all validator slots have been filled, marking a major milestone for UOMI’s Layer 1 blockchain. This rapid adoption underscores the strong demand for our technology and the growing recognition of our decentralized AI-powered ecosystem. UOMI it’s a cutting-edge L1 infrastructure designed to power on-chain autonomous AI Agents and push the boundaries of decentralized AI. With our advanced OPoC consensus mechanism, we’re building a scalable and efficient network that enables secure AI inference and computation while maintaining decentralization. This incredible validator adoption is a testament to the strength of our community and technology. And we’re just getting started! 🔜 What’s next? Stay tuned for major announcements, including new strategic partnerships, exclusive opportunities for users, and updates on the upcoming $UOMI token launch. The momentum is unstoppable, and we’re building at full speed! Thank you to everyone who joined us in this first phase the future of Web3 AI is here, and you’re part of it! 🚀
0:20

🔥 We’re excited to announce that the Finney Testnet is thriving! In record time, all validator slots have been filled, marking a major milestone for UOMI’s Layer 1 blockchain. This rapid adoption underscores the strong demand for our technology and the growing recognition of our decentralized AI-powered ecosystem. UOMI it’s a cutting-edge L1 infrastructure designed to power on-chain autonomous AI Agents and push the boundaries of decentralized AI. With our advanced OPoC consensus mechanism, we’re building a scalable and efficient network that enables secure AI inference and computation while maintaining decentralization. This incredible validator adoption is a testament to the strength of our community and technology. And we’re just getting started! 🔜 What’s next? Stay tuned for major announcements, including new strategic partnerships, exclusive opportunities for users, and updates on the upcoming $UOMI token launch. The momentum is unstoppable, and we’re building at full speed! Thank you to everyone who joined us in this first phase the future of Web3 AI is here, and you’re part of it! 🚀

Uomi

16,830 views • 1 year ago

If intelligence is the log of compute… it starts with a lot of compute! And that’s why we’re scaling our GPU fleet faster than anyone else. Just last year, we added over 2 gigawatts of new capacity – roughly the output of 2 nuclear power plants. And today we’re going further, announcing the world's most powerful AI datacenter, located in southeastern Wisconsin. Fairwater is a seamless cluster of hundreds of thousands of NVIDIA GB200s, connected by enough fiber to circle the Earth 4.5 times. It will deliver 10x the performance of the world’s fastest supercomputer today, enabling AI training and inference workloads at a level never before seen. For AI training workloads, you need compute at exponential scale. That’s why we designed the datacenter, GPU fleet, and network together as one integrated system. This ensures a single job can run from day 1 at exponential scale across thousands of GPUs. Fairwater uses a liquid-cooled closed-loop system for cooling GPUs that requires zero water for operations after construction. And we’re matching all of the energy that is consumed with renewable sources. And of course, it is just one of several similar sites we’re lighting up across our 70+ regions. We have multiple identical Fairwater datacenters under construction in other locations across the US, in addition to our AI infrastructure already deployed in over 100 datacenters around the world, powering model training, test-time compute, RL tuning, and real-time inference at global scale. Too often during times like this, people go with the current and only later wonder, how did we get here? With Fairwater, we're charting a new path: doing the hard engineering work, bringing compute, network, and storage into one highly scaled cluster, and designing closed-loop energy systems to meet real-world computing needs. And partnering with local communities to ensure it's thoughtfully done in a way that is sustainable, creates new jobs, and expands opportunity. We are thrilled to see this take hold in Wisconsin, and we are just getting started.
0:21

If intelligence is the log of compute… it starts with a lot of compute! And that’s why we’re scaling our GPU fleet faster than anyone else. Just last year, we added over 2 gigawatts of new capacity – roughly the output of 2 nuclear power plants. And today we’re going further, announcing the world's most powerful AI datacenter, located in southeastern Wisconsin. Fairwater is a seamless cluster of hundreds of thousands of NVIDIA GB200s, connected by enough fiber to circle the Earth 4.5 times. It will deliver 10x the performance of the world’s fastest supercomputer today, enabling AI training and inference workloads at a level never before seen. For AI training workloads, you need compute at exponential scale. That’s why we designed the datacenter, GPU fleet, and network together as one integrated system. This ensures a single job can run from day 1 at exponential scale across thousands of GPUs. Fairwater uses a liquid-cooled closed-loop system for cooling GPUs that requires zero water for operations after construction. And we’re matching all of the energy that is consumed with renewable sources. And of course, it is just one of several similar sites we’re lighting up across our 70+ regions. We have multiple identical Fairwater datacenters under construction in other locations across the US, in addition to our AI infrastructure already deployed in over 100 datacenters around the world, powering model training, test-time compute, RL tuning, and real-time inference at global scale. Too often during times like this, people go with the current and only later wonder, how did we get here? With Fairwater, we're charting a new path: doing the hard engineering work, bringing compute, network, and storage into one highly scaled cluster, and designing closed-loop energy systems to meet real-world computing needs. And partnering with local communities to ensure it's thoughtfully done in a way that is sustainable, creates new jobs, and expands opportunity. We are thrilled to see this take hold in Wisconsin, and we are just getting started.

Satya Nadella

2,022,693 views • 10 months ago

I'm running Llama 4 Maverick at 620 t/s! I'm living in the future! Honestly, a large language model running this fast is something straight out of a sci-fi movie. Speeds like this will enable a whole new world of applications that aren't possible today. For reference, GPT-4o, which is probably the most popular OpenAI model, runs between 60 and 110 t/s. The secret here: I'm not running AI at Meta's Llama 4 Maverick on a GPU. I'm using the SambaNova Cloud (my sponsor) and their custom SN40L chips. They are optimized from the ground up for running AI workflows. Right now, SambaNova Cloud runs DeepSeek, Qwen, Whisper, and the entire family of Llama models on these chips. You can check the speed of each of these models using SambaNova Cloud's Playground (see the attached video). It's completely free, and that's how I'm measuring their speeds. For example, I also tried DeepSeek R1 (the latest version from May) and, oh boy! DeepSeek R1 is a huge 671B parameter model. It's probably the best open reasoning model in the world, and it runs at 140 tokens per second! !!! Inference time on an SN40L is night and day from what you'll get from a GPU. Here is why this is big: If you are running an agentic workflow that uses multiple models simultaneously on a GPU, it will need to swap models in and out of memory (because not every model fits). A single SNL40 chip can simultaneously hold over 100 models (trillions of parameters) in memory. If you are using open models, try the SambaCloud API to see what lightning speed looks like. Here is how: 1. Create a free account at: 2. Check the QuickStart guide: If you try the playground, check the speed you're getting with Llama 4 and DeepSeek, and post the results below. I've seen much higher numbers than I posted here, so I'm curious to see whether geography affects the speed.
1:10

I'm running Llama 4 Maverick at 620 t/s! I'm living in the future! Honestly, a large language model running this fast is something straight out of a sci-fi movie. Speeds like this will enable a whole new world of applications that aren't possible today. For reference, GPT-4o, which is probably the most popular OpenAI model, runs between 60 and 110 t/s. The secret here: I'm not running AI at Meta's Llama 4 Maverick on a GPU. I'm using the SambaNova Cloud (my sponsor) and their custom SN40L chips. They are optimized from the ground up for running AI workflows. Right now, SambaNova Cloud runs DeepSeek, Qwen, Whisper, and the entire family of Llama models on these chips. You can check the speed of each of these models using SambaNova Cloud's Playground (see the attached video). It's completely free, and that's how I'm measuring their speeds. For example, I also tried DeepSeek R1 (the latest version from May) and, oh boy! DeepSeek R1 is a huge 671B parameter model. It's probably the best open reasoning model in the world, and it runs at 140 tokens per second! !!! Inference time on an SN40L is night and day from what you'll get from a GPU. Here is why this is big: If you are running an agentic workflow that uses multiple models simultaneously on a GPU, it will need to swap models in and out of memory (because not every model fits). A single SNL40 chip can simultaneously hold over 100 models (trillions of parameters) in memory. If you are using open models, try the SambaCloud API to see what lightning speed looks like. Here is how: 1. Create a free account at: 2. Check the QuickStart guide: If you try the playground, check the speed you're getting with Llama 4 and DeepSeek, and post the results below. I've seen much higher numbers than I posted here, so I'm curious to see whether geography affects the speed.

Santiago

34,148 views • 1 year ago

🎉 The best way to start the week is to find out that our MedSAM is finally published today in Nature Communications! **Segment anything in medical images** Paper: arXiv: Data & Code: MedSAM is the first promotable foundation model for medical image segmentation. **Highlights**: ⭐ Before its formal publication, we have received 220 citations and 1400+ GitHub stars 🙏🙏❤️‍🔥❤️‍🔥❤️‍🔥 📊 We curated a large-scale medical image dataset with 1,570,263 image-mask pairs, covering 10 imaging modalities and over 30 cancer types. 🚀 Built on top of SAM (AI at Meta ) with transfer learning, we have significantly enhanced its segmentation performance of medical images. 📈 Comprehensive evaluations of 86 internal validation tasks and 60 external validation tasks demonstrate its better accuracy and robustness than modality-wise specialist models. **What is Next? --- Clinical Translation!!** 🍕Our next goal is to make the model deployable on laptops (CPUs) or other edge devices without reliance on GPUs. We have distilled a lightweight model, LiteMedSAM, offering a speed boost of 10x while maintaining accuracy. Plus, we have integrated it into the 3D Slicer plugin, providing an efficient tool for medical image segmentation. 🌐 To further promote developments in this field, we organize a competition on #CVPR2026: Segment Anything in Medical Images on Laptop! An out-of-the-box baseline has been released to reduce the entry barriers. Welcome to join us to push the boundary further: 🙏 Massive thanks to MetaAI AI at Meta for their open-source project SAM and many reviewers/users for their invaluable feedback. A huge shoutout to my postdoc Jun Ma (JunMa) for his leadership on this project!! UHN AI Hub Vector Institute Peter Munk Cardiac Centre AI Department of Laboratory Medicine & Pathobiology U of T Department of Computer Science University of Toronto University Health Network Brad Wouters 🇨🇦 Barry Rubin MD, PhD, FRCSC Shaf Keshavjee
0:59

🎉 The best way to start the week is to find out that our MedSAM is finally published today in Nature Communications! **Segment anything in medical images** Paper: arXiv: Data & Code: MedSAM is the first promotable foundation model for medical image segmentation. **Highlights**: ⭐ Before its formal publication, we have received 220 citations and 1400+ GitHub stars 🙏🙏❤️‍🔥❤️‍🔥❤️‍🔥 📊 We curated a large-scale medical image dataset with 1,570,263 image-mask pairs, covering 10 imaging modalities and over 30 cancer types. 🚀 Built on top of SAM (AI at Meta ) with transfer learning, we have significantly enhanced its segmentation performance of medical images. 📈 Comprehensive evaluations of 86 internal validation tasks and 60 external validation tasks demonstrate its better accuracy and robustness than modality-wise specialist models. **What is Next? --- Clinical Translation!!** 🍕Our next goal is to make the model deployable on laptops (CPUs) or other edge devices without reliance on GPUs. We have distilled a lightweight model, LiteMedSAM, offering a speed boost of 10x while maintaining accuracy. Plus, we have integrated it into the 3D Slicer plugin, providing an efficient tool for medical image segmentation. 🌐 To further promote developments in this field, we organize a competition on #CVPR2026: Segment Anything in Medical Images on Laptop! An out-of-the-box baseline has been released to reduce the entry barriers. Welcome to join us to push the boundary further: 🙏 Massive thanks to MetaAI AI at Meta for their open-source project SAM and many reviewers/users for their invaluable feedback. A huge shoutout to my postdoc Jun Ma (JunMa) for his leadership on this project!! UHN AI Hub Vector Institute Peter Munk Cardiac Centre AI Department of Laboratory Medicine & Pathobiology U of T Department of Computer Science University of Toronto University Health Network Brad Wouters 🇨🇦 Barry Rubin MD, PhD, FRCSC Shaf Keshavjee

Bo Wang

140,208 views • 2 years ago

$TAO just reclaimed the #1 AI crypto spot. Most people saw the headline. Almost nobody understands what it means for the price. Here is the data. $NEAR built real infrastructure. Partnerships. Developer activity. A legitimate ecosystem. $TAO just walked past it anyway. Not because of hype. Because Bittensor is the only AI crypto with a functioning marketplace for machine intelligence where supply, demand, and price discovery are all happening on-chain right now. That is not a roadmap. That is a live network. The numbers. 120+ subnets running today. $1.4B+ total ecosystem value. Chutes AI subnet: 150B+ tokens per day. Grayscale GTAO Trust: already live. Single subnet listed on the marketplace at $970,000 asking price. Subnets are becoming assets. The market is starting to price that. What the emission data is telling you. Emission rate is the network's vote on where the most valuable work is being done. When a subnet gains emission share, the collective stake-weighted intelligence of the network has decided that subnet's output is worth more of the TAO supply. Chutes AI gaining emissions while processing 150B tokens daily is not a coincidence. The network is directing capital toward proven output before any headline announces it. Why mainstream money changes everything. James Altucher just launched bluetao. ai, a TAO-powered ChatGPT alternative built directly on Bittensor subnets. He did not just buy the token. He built a product on the network. Products built on a network create structural demand for the native asset. That is how every successful L1 cycle has worked. Bittensor is now getting that builder activity from outside the crypto native world. That is a different signal from a price target tweet. Why $TAO is structurally different. Most AI tokens are betting their chain becomes the preferred environment for AI development. $TAO is not betting on becoming infrastructure. It already is. 120+ subnets running. Miners competing. Validators setting weights. Alpha tokens being priced in real time. The difference between $TAO and every other AI crypto is the difference between a city under construction and a city people are already living in. Van de Poppe said $1,000 to $2,000 in 12 months. He gave you the narrative. The subnet emission data is the mechanism he did not explain. Now you have both. $TAO at $313 with a $3.42B market cap is still early relative to what this network is actually processing. Centralised AI infrastructure companies are valued at hundreds of billions for processing far less novel work than a decentralised intelligence marketplace running 120+ competing subnets simultaneously. The repricing has not happened yet. The subnet marketplace listing at $970,000 is telling you something the price has not caught up to yet.
16:52

$TAO just reclaimed the #1 AI crypto spot. Most people saw the headline. Almost nobody understands what it means for the price. Here is the data. $NEAR built real infrastructure. Partnerships. Developer activity. A legitimate ecosystem. $TAO just walked past it anyway. Not because of hype. Because Bittensor is the only AI crypto with a functioning marketplace for machine intelligence where supply, demand, and price discovery are all happening on-chain right now. That is not a roadmap. That is a live network. The numbers. 120+ subnets running today. $1.4B+ total ecosystem value. Chutes AI subnet: 150B+ tokens per day. Grayscale GTAO Trust: already live. Single subnet listed on the marketplace at $970,000 asking price. Subnets are becoming assets. The market is starting to price that. What the emission data is telling you. Emission rate is the network's vote on where the most valuable work is being done. When a subnet gains emission share, the collective stake-weighted intelligence of the network has decided that subnet's output is worth more of the TAO supply. Chutes AI gaining emissions while processing 150B tokens daily is not a coincidence. The network is directing capital toward proven output before any headline announces it. Why mainstream money changes everything. James Altucher just launched bluetao. ai, a TAO-powered ChatGPT alternative built directly on Bittensor subnets. He did not just buy the token. He built a product on the network. Products built on a network create structural demand for the native asset. That is how every successful L1 cycle has worked. Bittensor is now getting that builder activity from outside the crypto native world. That is a different signal from a price target tweet. Why $TAO is structurally different. Most AI tokens are betting their chain becomes the preferred environment for AI development. $TAO is not betting on becoming infrastructure. It already is. 120+ subnets running. Miners competing. Validators setting weights. Alpha tokens being priced in real time. The difference between $TAO and every other AI crypto is the difference between a city under construction and a city people are already living in. Van de Poppe said $1,000 to $2,000 in 12 months. He gave you the narrative. The subnet emission data is the mechanism he did not explain. Now you have both. $TAO at $313 with a $3.42B market cap is still early relative to what this network is actually processing. Centralised AI infrastructure companies are valued at hundreds of billions for processing far less novel work than a decentralised intelligence marketplace running 120+ competing subnets simultaneously. The repricing has not happened yet. The subnet marketplace listing at $970,000 is telling you something the price has not caught up to yet.

2xnmore

12,150 views • 2 months ago

🚨 SUBNET 35 BREAKTHROUGH: LogicNet AI Rivals HUMAN EXPERTS! 🧠💥@logicnet_ait Choose your weapon: Qwen2, Mistral, Baichuan2, or Falcon models now available! 🛠️ $TAO holders are part-owners of a GLOBAL AI BRAIN trust! As these models improve, so does the VALUE and UTILITY of the entire ecosystem 🔥 Bittensor's Subnet 35 isn't just improving AI - it's REVOLUTIONIZING it! Here's what these mind-blowing stats REALLY mean: 1. 📚 SAT Performance (53.1%): - This is like an AI scoring 1060 out of 1600 on the SAT! - Puts the AI at the 50th percentile of HUMAN test-takers - Could help MILLIONS of students with SAT prep! 2. 🔬 MMLU-STEM (63.9% by Minerva 540B): - MMLU tests grad-school level knowledge - This score rivals EXPERT HUMANS in STEM fields - Potential to assist in CUTTING-EDGE research and innovation 3. 🧮 Math Skills (33.6% on complex tasks): - Tackles problems at the level of Math Olympiad competitions - Could speed up COMPLEX CALCULATIONS in physics, engineering, and finance - Imagine AI co-pilots for researchers and engineers! 4. 💡 Logical Reasoning (36.4% on GSM8k): - GSM8k tests multi-step problem-solving - This level could help in LEGAL ANALYSIS, BUSINESS STRATEGY, and SCIENTIFIC RESEARCH - AI assistants that can break down complex problems for humans! 5. 🌐 Scale and Diversity (7B to 540B parameters): - From models that can run on your laptop to those rivaling TOP TECH GIANTS - Democratizes access to AI power for EVERYONE - Potential for personalized AI assistants tailored to YOUR needs 🚀 Real-World Impact: - Education: AI tutors available 24/7, personalizing learning for MILLIONS - Research: Accelerating discoveries in medicine, climate science, and beyond - Business: AI strategists helping startups compete with FORTUNE 500 companies - Technology: Pushing the boundaries of what's possible in AI, rivaling or SURPASSING centralized tech giants This isn't just progress - it's a LEAP in decentralized AI capabilities! Bittensor's Subnet 35 is creating AI that doesn't just assist humans, but COLLABORATES with us at expert levels! 🌟 Why This MATTERS for Bittensor & $TAO: - Decentralized AI now COMPETING with multi-billion dollar tech companies Don't just watch the future unfold - be part of SHAPING it with Bittensor & $TAO! 🚀🌍 Watch and Subcribe on YouTube for More: #Bittensor #TAO #AIRevolution #DecentralizedAI #FutureIsNow [DISCLAIMER: Not financial advice! DYOR. But remember, we're not just talking about numbers - we're talking about AI that could transform EVERY aspect of our lives! 🤯🚀]
1:22

🚨 SUBNET 35 BREAKTHROUGH: LogicNet AI Rivals HUMAN EXPERTS! 🧠💥@logicnet_ait Choose your weapon: Qwen2, Mistral, Baichuan2, or Falcon models now available! 🛠️ $TAO holders are part-owners of a GLOBAL AI BRAIN trust! As these models improve, so does the VALUE and UTILITY of the entire ecosystem 🔥 Bittensor's Subnet 35 isn't just improving AI - it's REVOLUTIONIZING it! Here's what these mind-blowing stats REALLY mean: 1. 📚 SAT Performance (53.1%): - This is like an AI scoring 1060 out of 1600 on the SAT! - Puts the AI at the 50th percentile of HUMAN test-takers - Could help MILLIONS of students with SAT prep! 2. 🔬 MMLU-STEM (63.9% by Minerva 540B): - MMLU tests grad-school level knowledge - This score rivals EXPERT HUMANS in STEM fields - Potential to assist in CUTTING-EDGE research and innovation 3. 🧮 Math Skills (33.6% on complex tasks): - Tackles problems at the level of Math Olympiad competitions - Could speed up COMPLEX CALCULATIONS in physics, engineering, and finance - Imagine AI co-pilots for researchers and engineers! 4. 💡 Logical Reasoning (36.4% on GSM8k): - GSM8k tests multi-step problem-solving - This level could help in LEGAL ANALYSIS, BUSINESS STRATEGY, and SCIENTIFIC RESEARCH - AI assistants that can break down complex problems for humans! 5. 🌐 Scale and Diversity (7B to 540B parameters): - From models that can run on your laptop to those rivaling TOP TECH GIANTS - Democratizes access to AI power for EVERYONE - Potential for personalized AI assistants tailored to YOUR needs 🚀 Real-World Impact: - Education: AI tutors available 24/7, personalizing learning for MILLIONS - Research: Accelerating discoveries in medicine, climate science, and beyond - Business: AI strategists helping startups compete with FORTUNE 500 companies - Technology: Pushing the boundaries of what's possible in AI, rivaling or SURPASSING centralized tech giants This isn't just progress - it's a LEAP in decentralized AI capabilities! Bittensor's Subnet 35 is creating AI that doesn't just assist humans, but COLLABORATES with us at expert levels! 🌟 Why This MATTERS for Bittensor & $TAO: - Decentralized AI now COMPETING with multi-billion dollar tech companies Don't just watch the future unfold - be part of SHAPING it with Bittensor & $TAO! 🚀🌍 Watch and Subcribe on YouTube for More: #Bittensor #TAO #AIRevolution #DecentralizedAI #FutureIsNow [DISCLAIMER: Not financial advice! DYOR. But remember, we're not just talking about numbers - we're talking about AI that could transform EVERY aspect of our lives! 🤯🚀]

Andy ττ

30,418 views • 1 year ago

Groq is serving the fastest responses I've ever seen. We're talking almost 500 T/s! I did some research on how they're able to do it. Turns out they developed their own hardware that utilize LPUs instead of GPUs. Here's the skinny: Groq created a novel processing unit known as the Tensor Streaming Processor (TSP) which they categorize as a Linear Processor Unit (LPU). Unlike traditional GPUs that are parallel processors with hundreds of cores designed for graphics rendering, LPUs are architected to deliver deterministic performance for AI computations. The LPU's architecture is a departure from the SIMD (Single Instruction, Multiple Data) model used by GPUs and favor a more streamlined approach that eliminate the need for complex scheduling hardware. This design allows every clock cycle to be utilized effectively, ensuring consistent latency and throughput. For developers, this means that performance can be precisely predicted and optimized which is critical in real-time AI applications. Energy efficiency is another area where LPUs shine. By reducing the overhead of managing multiple threads and avoiding the underutilization of cores, LPUs can deliver more computations per watt. Groq's innovative chip design allows multiple TSPs to be linked together without the traditional bottlenecks found in GPU clusters making them extremely scalable. This enables linear scaling of performance as more LPUs are added simplifying the hardware requirements for large-scale AI models and making it easier for developers to scale their applications without rearchitecting their systems. So what does this all mean? LPUs could provide a massive improvement compared to GPUs for serving AI applications in the future! If anything it will be great to have alternative high performing hardware since A100s and H100s are so in demand
0:48

Groq is serving the fastest responses I've ever seen. We're talking almost 500 T/s! I did some research on how they're able to do it. Turns out they developed their own hardware that utilize LPUs instead of GPUs. Here's the skinny: Groq created a novel processing unit known as the Tensor Streaming Processor (TSP) which they categorize as a Linear Processor Unit (LPU). Unlike traditional GPUs that are parallel processors with hundreds of cores designed for graphics rendering, LPUs are architected to deliver deterministic performance for AI computations. The LPU's architecture is a departure from the SIMD (Single Instruction, Multiple Data) model used by GPUs and favor a more streamlined approach that eliminate the need for complex scheduling hardware. This design allows every clock cycle to be utilized effectively, ensuring consistent latency and throughput. For developers, this means that performance can be precisely predicted and optimized which is critical in real-time AI applications. Energy efficiency is another area where LPUs shine. By reducing the overhead of managing multiple threads and avoiding the underutilization of cores, LPUs can deliver more computations per watt. Groq's innovative chip design allows multiple TSPs to be linked together without the traditional bottlenecks found in GPU clusters making them extremely scalable. This enables linear scaling of performance as more LPUs are added simplifying the hardware requirements for large-scale AI models and making it easier for developers to scale their applications without rearchitecting their systems. So what does this all mean? LPUs could provide a massive improvement compared to GPUs for serving AI applications in the future! If anything it will be great to have alternative high performing hardware since A100s and H100s are so in demand

Jay Scambler

318,228 views • 2 years ago

🚀 Exciting News $SHELL the Future of AI Agents and $TAO with the TTS Subnet on Bittensor! 🚀 MyShell extends the impact of Bittensor's incentive mechanism to its over 1 million registered users and 50,000 creators, greatly expanding Bittensor's and $TAO's influence. MyShell and Bittensor are right at the heart this massive shake-up with $SHELL and $TAO. They’re all about making AI not just smart but also something everyone can get into, thanks to the power of decentralized networks. And at the core? AI agents. These aren't your average digital assistants; they're about to change how we interact with tech on a whole new level. MyShell's Big Idea with $SHELL So, MyShell’s got this big plan to make AI something anyone can dive into. They're launching this TTS Subnet thing on Bittensor's network, which is all about making machines talk in more human-like ways, and they're using $SHELL tokens to fuel this vision. Their goal? To push past old-school AI limits and create with AI as easy as pie, all while keeping it open-source and community-powered. Bittensor Does Its Magic with $TAO On the other side, you’ve got Bittensor doing wonders with $TAO, building this massive network where anyone, anywhere, can chip in on AI research and development. This partnership with MyShell? It’s a game-changer, breaking down walls in AI development and letting folks from all over the world have a go at making AI smarter. $SHELL + $TAO = AI Revolution Putting $SHELL and $TAO together is where the magic really happens. MyShell and Bittensor aren’t just teaming up for the tech; they’re here to transform our digital world, making AI agents a big part of our online lives. Imagine AI that doesn’t just follow orders but helps, creates, and learns with you. That’s the future they’re building. Hop on Board the AI Revolution This isn’t just tech talk; it’s a call to action. MyShell and Bittensor are inviting anyone with a spark for AI to jump in and help shape this new world. Whether you’re a coder, a creator, or just curious, there’s a spot for you to dive in and make a difference. Want to get started? Check out MyShell on GitHub: Follow the latest buzz on X: MyShell.AI Take a deeper dive at Website: This is more than just building AI; it’s about crafting a future where AI is part of everyone’s life, powered by the community, for the community. Let’s make it happen with $SHELL and $TAO. Share on YouTube:
4:21

🚀 Exciting News $SHELL the Future of AI Agents and $TAO with the TTS Subnet on Bittensor! 🚀 MyShell extends the impact of Bittensor's incentive mechanism to its over 1 million registered users and 50,000 creators, greatly expanding Bittensor's and $TAO's influence. MyShell and Bittensor are right at the heart this massive shake-up with $SHELL and $TAO. They’re all about making AI not just smart but also something everyone can get into, thanks to the power of decentralized networks. And at the core? AI agents. These aren't your average digital assistants; they're about to change how we interact with tech on a whole new level. MyShell's Big Idea with $SHELL So, MyShell’s got this big plan to make AI something anyone can dive into. They're launching this TTS Subnet thing on Bittensor's network, which is all about making machines talk in more human-like ways, and they're using $SHELL tokens to fuel this vision. Their goal? To push past old-school AI limits and create with AI as easy as pie, all while keeping it open-source and community-powered. Bittensor Does Its Magic with $TAO On the other side, you’ve got Bittensor doing wonders with $TAO, building this massive network where anyone, anywhere, can chip in on AI research and development. This partnership with MyShell? It’s a game-changer, breaking down walls in AI development and letting folks from all over the world have a go at making AI smarter. $SHELL + $TAO = AI Revolution Putting $SHELL and $TAO together is where the magic really happens. MyShell and Bittensor aren’t just teaming up for the tech; they’re here to transform our digital world, making AI agents a big part of our online lives. Imagine AI that doesn’t just follow orders but helps, creates, and learns with you. That’s the future they’re building. Hop on Board the AI Revolution This isn’t just tech talk; it’s a call to action. MyShell and Bittensor are inviting anyone with a spark for AI to jump in and help shape this new world. Whether you’re a coder, a creator, or just curious, there’s a spot for you to dive in and make a difference. Want to get started? Check out MyShell on GitHub: Follow the latest buzz on X: MyShell.AI Take a deeper dive at Website: This is more than just building AI; it’s about crafting a future where AI is part of everyone’s life, powered by the community, for the community. Let’s make it happen with $SHELL and $TAO. Share on YouTube:

Andy ττ

10,823 views • 2 years ago

Today we announced our new Fairwater datacenter in Atlanta, connected with our first Fairwater site in Wisconsin and our broader Azure footprint to create the world’s first AI superfactory. Fairwater exemplifies our vision for a fungible fleet: infra that can serve any workload, anywhere, on fit-for-purpose accelerators and network paths, with maximum performance and efficiency. AI workloads have evolved beyond large-scale pre-training. Today, they encompass fine-tuning, reinforcement learning (RL), synthetic data generation, evaluation pipelines, and more. Fairwater is built to support this full lifecycle: Max density: Fairwater’s two-story design and liquid cooling system lets us place racks in three dimensions and pack them with GPUs as densely as possible, minimizing cable runs and improving latency and effective bandwidth. Fleet: Each Fairwater DC can integrate hundreds of thousands of the latest NVIDIA GPUs into a single coherent cluster. This provides flexible infra that can support the full spectrum of workloads, and ensure no GPU is left unnecessarily idle. And that’s on top of the more than 100,000 GB300s coming online this quarter alone for inference across the rest of our fleet. For us, it’s all about turning every gigawatt into the maximum number of useful tokens. Not every GW is created equal! Planet-scale: Every Fairwater DC will connect through our continent-spanning AI WAN to prior generations of AI supercomputers, forming a truly fungible pool of compute. This enables developers to scale beyond the capacity of a single site and dynamically land workloads on the right infra for their needs. Together, these innovations let us bring together different generations of silicon and AI systems across DCs and geos into a single elastic system that scales seamlessly across training and inference workloads And this elastic AI capacity is all available alongside all the other cloud services (compute, storage, databases, app services) that AI agents and workloads need. This is what we mean when we talk about building a fungible fleet – a single, unified platform that pushes the limits of performance per watt and per dollar. Read more:
0:44

Today we announced our new Fairwater datacenter in Atlanta, connected with our first Fairwater site in Wisconsin and our broader Azure footprint to create the world’s first AI superfactory. Fairwater exemplifies our vision for a fungible fleet: infra that can serve any workload, anywhere, on fit-for-purpose accelerators and network paths, with maximum performance and efficiency. AI workloads have evolved beyond large-scale pre-training. Today, they encompass fine-tuning, reinforcement learning (RL), synthetic data generation, evaluation pipelines, and more. Fairwater is built to support this full lifecycle: Max density: Fairwater’s two-story design and liquid cooling system lets us place racks in three dimensions and pack them with GPUs as densely as possible, minimizing cable runs and improving latency and effective bandwidth. Fleet: Each Fairwater DC can integrate hundreds of thousands of the latest NVIDIA GPUs into a single coherent cluster. This provides flexible infra that can support the full spectrum of workloads, and ensure no GPU is left unnecessarily idle. And that’s on top of the more than 100,000 GB300s coming online this quarter alone for inference across the rest of our fleet. For us, it’s all about turning every gigawatt into the maximum number of useful tokens. Not every GW is created equal! Planet-scale: Every Fairwater DC will connect through our continent-spanning AI WAN to prior generations of AI supercomputers, forming a truly fungible pool of compute. This enables developers to scale beyond the capacity of a single site and dynamically land workloads on the right infra for their needs. Together, these innovations let us bring together different generations of silicon and AI systems across DCs and geos into a single elastic system that scales seamlessly across training and inference workloads And this elastic AI capacity is all available alongside all the other cloud services (compute, storage, databases, app services) that AI agents and workloads need. This is what we mean when we talk about building a fungible fleet – a single, unified platform that pushes the limits of performance per watt and per dollar. Read more:

Satya Nadella

907,893 views • 8 months ago

The Cost of Intelligence is Heading to Zero | Hyperspace P2P Distributed Cache We present to you our breakthrough cross-domain work across AI, distributed systems, cryptography, game theory to solve the primary structural inefficiency at the heart of AI infrastructure: most inference is redundant. Google has reported that only 15% of daily searches are truly novel. The rest are repeats or close variants. LLM inference inherits this same power-law distribution. Enterprise chatbots see 70-80% of queries fall into a handful of intent categories. System prompts are identical across 100% of requests within an application. The KV attention state for "You are a helpful assistant" has been computed billions of times, on millions of GPUs, identically. And yet every AI lab, every startup, every self-hosted deployment - computes and caches these results independently. There is no shared layer. No global memory. Every provider pays the full compute cost for every query, even when the answer already exists somewhere in the network. This is the problem Hyperspace solves where distributed cache operates at three levels, each catching a different class of redundancy: 1. Response cache Same prompt, same model, same parameters - instant cached response from any node in the network. SHA-256 hash lookup via DHT, with cryptographic cache proofs linking every response to its original inference execution. No trust required. Fetchers re-announce as providers, so popular responses replicate naturally across more nodes. 2. KV prefix cache Same system prompt tokens - skip the most expensive part of inference entirely. Prefill (computing Key-Value attention states) is deterministic: same model plus same tokens always produces identical KV state. The network caches these states using erasure coding and distributes them via the routing network. New questions that share a common prefix resume generation from cached state instead of recomputing from scratch. 3. Routing to cached nodes Instead of transferring KV state across the network for every request, Hyperspace routes the request to the node that already has the state loaded in VRAM. The request goes to the cache, not the cache to the request. Together, these three layers mean that 70-90% of inference requests at network scale never require full GPU computation. This work doesn't exist in isolation. It builds on research from across the industry: SGLang's RadixAttention demonstrated that automatic prefix sharing can yield up to 5x speedup on structured LLM workloads. Moonshot AI's Mooncake built an entire KV-cache-centric disaggregated architecture for production serving at Kimi. Anthropic, OpenAI, and Google all launched prompt caching products in 2024 - priced at 50-90% discounts - because system prompt reuse is so pervasive that it changes the economics of inference. What all of these systems share is a common limitation: they operate within a single organization's infrastructure. SGLang caches prefixes within one server. Mooncake disaggregates KV cache within one datacenter. Anthropic's prompt caching works within one API provider's fleet. None of them can share cached state across organizational boundaries. Hyperspace removes this boundary. The cache is global. A response computed by a node in Tokyo is immediately available to a node in Berlin. A KV prefix state generated for Qwen-32B on one machine is verifiable and reusable by any other machine running the same model. The routing network provides the delivery guarantees, the erasure coding provides the redundancy, and the cache proofs provide the trust. What this means for the cost of intelligence Big AI labs scale linearly: twice the users means twice the GPU spend. Every query is a cost center. Their internal caching helps, but it's siloed - Lab A's cache can't serve Lab B's users, and neither can serve a self-hosted Llama deployment. Hyperspace scales sub-linearly. Every new node that joins the network adds to the global cache. Every inference result enriches the cache for all future requests. The cache hit rate rises with network size because query distributions follow a power law - the most common questions are asked exponentially more often than rare ones. The implication is simple: as the network grows, the effective cost per inference drops. Not linearly. Logarithmically. At 10 million nodes, we estimate 75-90% of all inference requests can be served from cache, eliminating 400,000+ MWh of energy consumption per year and avoiding over 200,000 tons of CO2 emissions. The first person to ask a question pays the compute cost. Everyone after them gets the answer for free, with cryptographic proof that it's authentic. Training is competitive. Inference is shared Open-weight models are converging on quality with closed models. Labs will continue to differentiate on training - data curation, architecture innovation, RLHF tuning. That's where the real intellectual property lives. But inference is a commodity. Two copies of Qwen-32B running the same prompt produce the same KV state and the same response, byte for byte, regardless of whose GPU runs the matrix multiplication. There is no moat in multiplying matrices. The moat is in training the weights. A global distributed cache makes this separation explicit. It doesn't matter who trained the model. Once the weights are open, the inference cost approaches zero at scale - because the network remembers every answer and can prove it's correct. No lab, no matter how well-funded, can match this. They cannot share caches across competitors. They scale linearly. The network scales logarithmically. The marginal cost of intelligence approaches zero. That's the endgame.
0:23

The Cost of Intelligence is Heading to Zero | Hyperspace P2P Distributed Cache We present to you our breakthrough cross-domain work across AI, distributed systems, cryptography, game theory to solve the primary structural inefficiency at the heart of AI infrastructure: most inference is redundant. Google has reported that only 15% of daily searches are truly novel. The rest are repeats or close variants. LLM inference inherits this same power-law distribution. Enterprise chatbots see 70-80% of queries fall into a handful of intent categories. System prompts are identical across 100% of requests within an application. The KV attention state for "You are a helpful assistant" has been computed billions of times, on millions of GPUs, identically. And yet every AI lab, every startup, every self-hosted deployment - computes and caches these results independently. There is no shared layer. No global memory. Every provider pays the full compute cost for every query, even when the answer already exists somewhere in the network. This is the problem Hyperspace solves where distributed cache operates at three levels, each catching a different class of redundancy: 1. Response cache Same prompt, same model, same parameters - instant cached response from any node in the network. SHA-256 hash lookup via DHT, with cryptographic cache proofs linking every response to its original inference execution. No trust required. Fetchers re-announce as providers, so popular responses replicate naturally across more nodes. 2. KV prefix cache Same system prompt tokens - skip the most expensive part of inference entirely. Prefill (computing Key-Value attention states) is deterministic: same model plus same tokens always produces identical KV state. The network caches these states using erasure coding and distributes them via the routing network. New questions that share a common prefix resume generation from cached state instead of recomputing from scratch. 3. Routing to cached nodes Instead of transferring KV state across the network for every request, Hyperspace routes the request to the node that already has the state loaded in VRAM. The request goes to the cache, not the cache to the request. Together, these three layers mean that 70-90% of inference requests at network scale never require full GPU computation. This work doesn't exist in isolation. It builds on research from across the industry: SGLang's RadixAttention demonstrated that automatic prefix sharing can yield up to 5x speedup on structured LLM workloads. Moonshot AI's Mooncake built an entire KV-cache-centric disaggregated architecture for production serving at Kimi. Anthropic, OpenAI, and Google all launched prompt caching products in 2024 - priced at 50-90% discounts - because system prompt reuse is so pervasive that it changes the economics of inference. What all of these systems share is a common limitation: they operate within a single organization's infrastructure. SGLang caches prefixes within one server. Mooncake disaggregates KV cache within one datacenter. Anthropic's prompt caching works within one API provider's fleet. None of them can share cached state across organizational boundaries. Hyperspace removes this boundary. The cache is global. A response computed by a node in Tokyo is immediately available to a node in Berlin. A KV prefix state generated for Qwen-32B on one machine is verifiable and reusable by any other machine running the same model. The routing network provides the delivery guarantees, the erasure coding provides the redundancy, and the cache proofs provide the trust. What this means for the cost of intelligence Big AI labs scale linearly: twice the users means twice the GPU spend. Every query is a cost center. Their internal caching helps, but it's siloed - Lab A's cache can't serve Lab B's users, and neither can serve a self-hosted Llama deployment. Hyperspace scales sub-linearly. Every new node that joins the network adds to the global cache. Every inference result enriches the cache for all future requests. The cache hit rate rises with network size because query distributions follow a power law - the most common questions are asked exponentially more often than rare ones. The implication is simple: as the network grows, the effective cost per inference drops. Not linearly. Logarithmically. At 10 million nodes, we estimate 75-90% of all inference requests can be served from cache, eliminating 400,000+ MWh of energy consumption per year and avoiding over 200,000 tons of CO2 emissions. The first person to ask a question pays the compute cost. Everyone after them gets the answer for free, with cryptographic proof that it's authentic. Training is competitive. Inference is shared Open-weight models are converging on quality with closed models. Labs will continue to differentiate on training - data curation, architecture innovation, RLHF tuning. That's where the real intellectual property lives. But inference is a commodity. Two copies of Qwen-32B running the same prompt produce the same KV state and the same response, byte for byte, regardless of whose GPU runs the matrix multiplication. There is no moat in multiplying matrices. The moat is in training the weights. A global distributed cache makes this separation explicit. It doesn't matter who trained the model. Once the weights are open, the inference cost approaches zero at scale - because the network remembers every answer and can prove it's correct. No lab, no matter how well-funded, can match this. They cannot share caches across competitors. They scale linearly. The network scales logarithmically. The marginal cost of intelligence approaches zero. That's the endgame.

Varun

37,542 views • 4 months ago

Scale alone is not enough for AI data. Quality and complexity are equally critical. Excited to support all of these for LLM developers with Snorkel AI Data-as-a-Service, and to share our new leaderboard! — Our decade-plus of research and work in AI data has a simple point: scale alone is not enough. AI success is all about the quality, complexity, and distribution of data—in addition to volume. We’re excited to be powering leading LLM developers with Snorkel AI Expert Data-as-a-Service, our white glove service for custom, expert-level AI datasets—and to now preview some of what we’re building via our new Expert Data Leaderboard (🔗 in 🧵) + upcoming OSS dataset releases! Snorkel Expert Data-as-a-Service is built to meet the rapidly evolving data needs of the agentic AI world—where success is built on the quality, complexity, and distribution of datasets, in addition to size and scale. This kind of high-quality, frontier AI data can only come from a union of technology and human expertise. With Snorkel Expert Data-as-a-Service, we’re powering frontier LLM developers across agentic, expert knowledge, reasoning, coding, multi-modal, and other task types via the combination of these two key components: - (1) The Snorkel Expert Network: A global team of subject matter experts focused wholly on specialized knowledge–spanning thousands of topics in STEM/academic, vertical/professional, and consumer/lifestyle domains. - (2) Snorkel AI Data Development Platform: Our unique programmatic data curation and quality control platform, accelerating and improving expert authoring and review through principled techniques developed over the last decade of R&D. Now: we’re incredibly excited to showcase some of the power of Snorkel Expert Data-as-a-Service via the new Snorkel Leaderboard—putting frontier models to the test in complex, agentic, and reasoning settings inspired by real industry scenarios (not esoteric puzzles)! We’ll be releasing new leaderboards and accompanying expert-verified open source datasets (coming soon!) regularly. To start, we’re sharing three initial ones in preview: - SnorkelFinance: Q&A over financial documents requiring agentic tool-calling and reasoning - SnorkelUnderwrite: Agentic insurance tasks requiring industry-specific reasoning and tool use - SnorkelSequences: Mathematical tasks requiring compositional multi-step reasoning
1:11

Scale alone is not enough for AI data. Quality and complexity are equally critical. Excited to support all of these for LLM developers with Snorkel AI Data-as-a-Service, and to share our new leaderboard! — Our decade-plus of research and work in AI data has a simple point: scale alone is not enough. AI success is all about the quality, complexity, and distribution of data—in addition to volume. We’re excited to be powering leading LLM developers with Snorkel AI Expert Data-as-a-Service, our white glove service for custom, expert-level AI datasets—and to now preview some of what we’re building via our new Expert Data Leaderboard (🔗 in 🧵) + upcoming OSS dataset releases! Snorkel Expert Data-as-a-Service is built to meet the rapidly evolving data needs of the agentic AI world—where success is built on the quality, complexity, and distribution of datasets, in addition to size and scale. This kind of high-quality, frontier AI data can only come from a union of technology and human expertise. With Snorkel Expert Data-as-a-Service, we’re powering frontier LLM developers across agentic, expert knowledge, reasoning, coding, multi-modal, and other task types via the combination of these two key components: - (1) The Snorkel Expert Network: A global team of subject matter experts focused wholly on specialized knowledge–spanning thousands of topics in STEM/academic, vertical/professional, and consumer/lifestyle domains. - (2) Snorkel AI Data Development Platform: Our unique programmatic data curation and quality control platform, accelerating and improving expert authoring and review through principled techniques developed over the last decade of R&D. Now: we’re incredibly excited to showcase some of the power of Snorkel Expert Data-as-a-Service via the new Snorkel Leaderboard—putting frontier models to the test in complex, agentic, and reasoning settings inspired by real industry scenarios (not esoteric puzzles)! We’ll be releasing new leaderboards and accompanying expert-verified open source datasets (coming soon!) regularly. To start, we’re sharing three initial ones in preview: - SnorkelFinance: Q&A over financial documents requiring agentic tool-calling and reasoning - SnorkelUnderwrite: Agentic insurance tasks requiring industry-specific reasoning and tool use - SnorkelSequences: Mathematical tasks requiring compositional multi-step reasoning

Alex Ratner

495,851 views • 1 year ago

Researchers found a way to make LLMs 8.5x faster! (without compromising accuracy) Speculative decoding is quite an effective way to address the single-token bottleneck in traditional LLM inference. A small "draft" model first generates the next several tokens, then the large model verifies all of them at once in a single forward pass. If a token at any position is wrong, you keep everything before it and restart from there. This never does worse than normal decoding. But current drafters in Speculative decoding still guess one token at a time. That makes the drafting step itself a bottleneck, capping real-world speedups at 2-3x. DFlash is a new technique that swaps the autoregressive drafter with a lightweight block diffusion model that guesses all tokens in one parallel shot. Drafting cost stays flat no matter how many tokens you speculate. On top of that, the drafter is conditioned on hidden features pulled from multiple layers of the target model and injected into every draft layer, so it makes significantly better guesses than a drafter working from scratch. In the side-by-side demo below, vanilla decoding runs at 48.5 tokens/sec. DFlash hits 415 tokens/sec on the same model, with zero quality loss. It's already integrated with vLLM, SGLang, and Transformers, with draft models on HuggingFace for several models like Qwen3, Qwen3.5, Llama 3.1, Kimi-K2.5, gpt-oss, and many more. I have shared the GitHub repo in the replies! KV caching is another must-know technique to boost LLM inference. I recently wrote an article about it. Read it below. 👉 Over to you: What use case are you working on that can benefit from this new technique?
0:26

Researchers found a way to make LLMs 8.5x faster! (without compromising accuracy) Speculative decoding is quite an effective way to address the single-token bottleneck in traditional LLM inference. A small "draft" model first generates the next several tokens, then the large model verifies all of them at once in a single forward pass. If a token at any position is wrong, you keep everything before it and restart from there. This never does worse than normal decoding. But current drafters in Speculative decoding still guess one token at a time. That makes the drafting step itself a bottleneck, capping real-world speedups at 2-3x. DFlash is a new technique that swaps the autoregressive drafter with a lightweight block diffusion model that guesses all tokens in one parallel shot. Drafting cost stays flat no matter how many tokens you speculate. On top of that, the drafter is conditioned on hidden features pulled from multiple layers of the target model and injected into every draft layer, so it makes significantly better guesses than a drafter working from scratch. In the side-by-side demo below, vanilla decoding runs at 48.5 tokens/sec. DFlash hits 415 tokens/sec on the same model, with zero quality loss. It's already integrated with vLLM, SGLang, and Transformers, with draft models on HuggingFace for several models like Qwen3, Qwen3.5, Llama 3.1, Kimi-K2.5, gpt-oss, and many more. I have shared the GitHub repo in the replies! KV caching is another must-know technique to boost LLM inference. I recently wrote an article about it. Read it below. 👉 Over to you: What use case are you working on that can benefit from this new technique?

Avi Chawla

157,390 views • 2 months ago

Most $TAO holders are flying blind. They bought the token. They watched the price. They read the threads. But they have never opened the one tool that shows them everything happening inside the Bittensor network in real time. It is called Taostats. It is free. And after reading this, you will never look at $TAO the same way again. Here is exactly how to use it. Step 1: Start at the Subnets page. This is the heartbeat of the entire network. Every subnet running on Bittensor is listed here with: - its current emission rate - the number of active miners and validators - real-time performance data The emission rate is the most important number on this page. It tells you exactly how much TAO is flowing into each subnet every block. High emission means the network is directing significant resources toward that subnet's commodity. Low emission means the market has not yet recognised its value, or the subnet has not yet proven itself. Watch which subnets are gaining emission share over time. That movement tells you where the network believes the most valuable work is being done, before any headline announces it. Step 2: Use the Subnet pages to go deeper. Click any subnet, and you enter a complete dashboard for that individual market. - The TradingView chart shows you the alpha token price history for that subnet. Alpha tokens are the subnet-specific tokens that sit inside TAO's broader economy. Their price relative to TAO tells you how the market is valuing that subnet's specific commodity. - The Metagraph is the full list of every miner and validator currently active in the subnet: their UID, their stake, their trust score, their emission share. This is the raw intelligence layer. The miners consistently earning the most emissions are producing the work the validators collectively agree is the most valuable. - The Sentiment Index gives you a real-time community temperature reading on each subnet. Not price sentiment. Ecosystem sentiment. Whether the participants building inside the subnet believe it is healthy and improving. Step 3: Check Validators before you stake anything. This is the step most people skip and regret. The Validators page on Taostats shows you the performance history of every validator on the network: their VTrust score, their emission consistency, and their weight-setting behaviour across subnets. VTrust is the metric that matters most. It measures how closely a validator's judgments align with the honest stake-weighted majority across the network. High VTrust means the validator is doing genuine work and being rewarded for it. Low VTrust means the validator is either lazy, copying other validators' weights, or attempting to manipulate the system. When you delegate your TAO to a validator, you are trusting them with your emissions. Taostats shows you exactly which validators have earned that trust over time, and which ones have not. Never stake blind again. Step 4: Use the Blockchain explorer to track real movement. The Blockchain section of Taostats logs every transfer, every staking transaction, and every extrinsic called on the Bittensor chain in real time. This is where you track what wallets are actually doing: - Large staking transactions from unknown addresses - Subnet registration events that signal a new market is about to go live - Neuron registration burns that show demand for participation in a specific subnet is accelerating The people who read on-chain data before the narrative catches up to it are the ones who position correctly before the crowd notices the move. Step 5: Track your own portfolio inside the Dashboard. Connect your coldkey address, and Taostats builds you a complete portfolio view: - Your TAO balance - Your staking positions - Your delegation returns - Your yield over time The yield calculator is particularly useful. It shows you the actual return you are generating from your staking position in real TAO terms, not in percentage estimates that assume conditions that may not hold. If your yield is lower than the network average for your validator tier, Taostats shows you that too. Switching validators takes one transaction. The data to make that decision intelligently is right in front of you. The bigger picture. Most people holding $TAO are making decisions based on price charts and social media sentiment. Both of those inputs are downstream of what is actually happening inside the network. Subnet emission shifts. Validator VTrust changes. On-chain registration events. Neuron burn rates. Alpha token price movements relative to TAO. All of it is live on Taostats right now. All of it is free. All of it tells you something the price chart cannot. The investors who understand Bittensor at the data layer will always be positioned ahead of the investors who understand it at the narrative layer. Taostats is the data layer. Bookmark it. Open it daily. The network is telling you exactly what it is doing if you know where to look.
1:52

Most $TAO holders are flying blind. They bought the token. They watched the price. They read the threads. But they have never opened the one tool that shows them everything happening inside the Bittensor network in real time. It is called Taostats. It is free. And after reading this, you will never look at $TAO the same way again. Here is exactly how to use it. Step 1: Start at the Subnets page. This is the heartbeat of the entire network. Every subnet running on Bittensor is listed here with: - its current emission rate - the number of active miners and validators - real-time performance data The emission rate is the most important number on this page. It tells you exactly how much TAO is flowing into each subnet every block. High emission means the network is directing significant resources toward that subnet's commodity. Low emission means the market has not yet recognised its value, or the subnet has not yet proven itself. Watch which subnets are gaining emission share over time. That movement tells you where the network believes the most valuable work is being done, before any headline announces it. Step 2: Use the Subnet pages to go deeper. Click any subnet, and you enter a complete dashboard for that individual market. - The TradingView chart shows you the alpha token price history for that subnet. Alpha tokens are the subnet-specific tokens that sit inside TAO's broader economy. Their price relative to TAO tells you how the market is valuing that subnet's specific commodity. - The Metagraph is the full list of every miner and validator currently active in the subnet: their UID, their stake, their trust score, their emission share. This is the raw intelligence layer. The miners consistently earning the most emissions are producing the work the validators collectively agree is the most valuable. - The Sentiment Index gives you a real-time community temperature reading on each subnet. Not price sentiment. Ecosystem sentiment. Whether the participants building inside the subnet believe it is healthy and improving. Step 3: Check Validators before you stake anything. This is the step most people skip and regret. The Validators page on Taostats shows you the performance history of every validator on the network: their VTrust score, their emission consistency, and their weight-setting behaviour across subnets. VTrust is the metric that matters most. It measures how closely a validator's judgments align with the honest stake-weighted majority across the network. High VTrust means the validator is doing genuine work and being rewarded for it. Low VTrust means the validator is either lazy, copying other validators' weights, or attempting to manipulate the system. When you delegate your TAO to a validator, you are trusting them with your emissions. Taostats shows you exactly which validators have earned that trust over time, and which ones have not. Never stake blind again. Step 4: Use the Blockchain explorer to track real movement. The Blockchain section of Taostats logs every transfer, every staking transaction, and every extrinsic called on the Bittensor chain in real time. This is where you track what wallets are actually doing: - Large staking transactions from unknown addresses - Subnet registration events that signal a new market is about to go live - Neuron registration burns that show demand for participation in a specific subnet is accelerating The people who read on-chain data before the narrative catches up to it are the ones who position correctly before the crowd notices the move. Step 5: Track your own portfolio inside the Dashboard. Connect your coldkey address, and Taostats builds you a complete portfolio view: - Your TAO balance - Your staking positions - Your delegation returns - Your yield over time The yield calculator is particularly useful. It shows you the actual return you are generating from your staking position in real TAO terms, not in percentage estimates that assume conditions that may not hold. If your yield is lower than the network average for your validator tier, Taostats shows you that too. Switching validators takes one transaction. The data to make that decision intelligently is right in front of you. The bigger picture. Most people holding $TAO are making decisions based on price charts and social media sentiment. Both of those inputs are downstream of what is actually happening inside the network. Subnet emission shifts. Validator VTrust changes. On-chain registration events. Neuron burn rates. Alpha token price movements relative to TAO. All of it is live on Taostats right now. All of it is free. All of it tells you something the price chart cannot. The investors who understand Bittensor at the data layer will always be positioned ahead of the investors who understand it at the narrative layer. Taostats is the data layer. Bookmark it. Open it daily. The network is telling you exactly what it is doing if you know where to look.

2xnmore

141,195 views • 2 months ago

Dr. Fan video translation: First of all, I would like to congratulate more than 16 million Pioneers around the world for transitioning to the open network. This is the result of our joint efforts over the past six years. We should celebrate this historic moment in the development of the network. Looking back at the development of Pi now, it has always been a unique project. Some people will feel that it is different, and yes, it is. Pi is a non-conformist. Many innovations come from non-conformists who challenge conventions, conventions, and established practices. In the case of Pi, in the early days of cryptocurrency development, many projects raised funds through initial coin offerings (ICOs), but Pi never did so. Pi never sold tokens through ICOs and ensured that everyone could get them for free. 80% of Pi tokens belong to the public and the community. When a cryptocurrency project can go online with just a white paper, a smart contract, or even just an emoji today, the Pi community spent six years building its infrastructure, ecosystem, and ensuring that it has usable practical functions. Before opening, most of its addresses were unverified for most blockchains. Pi chooses to verify the identities of millions of users through identity verification (KYC) and business identity verification (KYB) to ensure that the identities of individuals and businesses on the mainnet are authentic and reliable. Because Pi firmly believes that true decentralization is not inconsistent with authenticity and legitimacy. Some cryptocurrency critics worry that Pi's users are too mainstream. Is this a problem? We think not. This is precisely the biggest advantage that the Pi network has been working hard to build and build over the past six years. Ordinary people like Pioneers are the driving force behind Pi's development because Pi aims to solve the problems of large-scale applications and real-world practicality. Pi welcomes both cryptocurrency enthusiasts and mainstream people. History has shown that only by meeting the real needs of the mainstream population can technology develop for a long time. With this development trend, the cryptocurrency industry as a whole will also benefit from the participation of a large number of mainstream audiences to jointly create the real utility of blockchain technology. What does it mean to open the network today? This means that the firewall of the Pi blockchain has been removed and connected to the outside world. This makes it easier for merchants to sell goods and services in the local market; developers can further develop applications and improve business model logic; creators can gain greater influence; and ordinary Pioneers can also connect and trade with each other. For new developers, welcome to this widely distributed crypto network. If you already have a business model, come to the Pi network to test and sell your products. For developers who don't have a business model but have an excellent application experience, the Pi application network has prepared a business model for you. The platform will bring you traffic and make you profitable. At the same time, it creates the use value of Pi for all pioneers, which is very beneficial to the Pi network as a whole. So, Pioneers, in the end, don't let external noise distract you. Focus on key matters, focus on things that can have an impact in the crypto field and even the world. Keep working hard, keep creating, and everything else will fall into place. MY MESSAGE Please listen carefully to this video so that you understand, the value is determined by the existing ecosystem, the current ecosystem is GCV $314,159 What small traders do under the auspices of the ambassadors 🥰 Eagle woman 🦅 Nonny Padja NTT 🇮🇩 Believe it or not, it's up to you, the point is we've won with a GCV barter value of $314,159 If you want to succeed Barter with Pioneer and small merchants with GCV value of $314,159❤️
4:38

Dr. Fan video translation: First of all, I would like to congratulate more than 16 million Pioneers around the world for transitioning to the open network. This is the result of our joint efforts over the past six years. We should celebrate this historic moment in the development of the network. Looking back at the development of Pi now, it has always been a unique project. Some people will feel that it is different, and yes, it is. Pi is a non-conformist. Many innovations come from non-conformists who challenge conventions, conventions, and established practices. In the case of Pi, in the early days of cryptocurrency development, many projects raised funds through initial coin offerings (ICOs), but Pi never did so. Pi never sold tokens through ICOs and ensured that everyone could get them for free. 80% of Pi tokens belong to the public and the community. When a cryptocurrency project can go online with just a white paper, a smart contract, or even just an emoji today, the Pi community spent six years building its infrastructure, ecosystem, and ensuring that it has usable practical functions. Before opening, most of its addresses were unverified for most blockchains. Pi chooses to verify the identities of millions of users through identity verification (KYC) and business identity verification (KYB) to ensure that the identities of individuals and businesses on the mainnet are authentic and reliable. Because Pi firmly believes that true decentralization is not inconsistent with authenticity and legitimacy. Some cryptocurrency critics worry that Pi's users are too mainstream. Is this a problem? We think not. This is precisely the biggest advantage that the Pi network has been working hard to build and build over the past six years. Ordinary people like Pioneers are the driving force behind Pi's development because Pi aims to solve the problems of large-scale applications and real-world practicality. Pi welcomes both cryptocurrency enthusiasts and mainstream people. History has shown that only by meeting the real needs of the mainstream population can technology develop for a long time. With this development trend, the cryptocurrency industry as a whole will also benefit from the participation of a large number of mainstream audiences to jointly create the real utility of blockchain technology. What does it mean to open the network today? This means that the firewall of the Pi blockchain has been removed and connected to the outside world. This makes it easier for merchants to sell goods and services in the local market; developers can further develop applications and improve business model logic; creators can gain greater influence; and ordinary Pioneers can also connect and trade with each other. For new developers, welcome to this widely distributed crypto network. If you already have a business model, come to the Pi network to test and sell your products. For developers who don't have a business model but have an excellent application experience, the Pi application network has prepared a business model for you. The platform will bring you traffic and make you profitable. At the same time, it creates the use value of Pi for all pioneers, which is very beneficial to the Pi network as a whole. So, Pioneers, in the end, don't let external noise distract you. Focus on key matters, focus on things that can have an impact in the crypto field and even the world. Keep working hard, keep creating, and everything else will fall into place. MY MESSAGE Please listen carefully to this video so that you understand, the value is determined by the existing ecosystem, the current ecosystem is GCV $314,159 What small traders do under the auspices of the ambassadors 🥰 Eagle woman 🦅 Nonny Padja NTT 🇮🇩 Believe it or not, it's up to you, the point is we've won with a GCV barter value of $314,159 If you want to succeed Barter with Pioneer and small merchants with GCV value of $314,159❤️

NONNY PADJA NTT ❤ Eagle woman 🦅

45,669 views • 1 year ago

What a year. 🚀 2025 was the year ChainOpera AI turned vision into real momentum: building a community-co-created, community-co-owned AI agent network and pushing the boundaries of what decentralized, collaborative intelligence can look like. 🚀 Biggest highlights from 2025 ✅- AI Terminal officially launched: We unveiled the ChainOpera AI Terminal as a unified gateway to decentralized AI, making it possible for anyone to interact with powerful, decentralized LLMs without technical friction. Positioned as the “browser for the DeAI era,” the AI Terminal marked a major step toward making decentralized intelligence accessible, usable, and mainstream. ✅- AI Terminal adoption at massive scale: Momentum followed quickly. The AI Terminal surpassed 2M registered users and consistently ranked top 3 among all apps on the BNB AI DappBay, validating strong product–market fit and real, sustained usage at scale. ✅- Announcing Coco: the world’s first community-owned Super Agent: We introduced Coco, the intelligence layer that sits between users and the agent network. Coco dynamically routes each request to the most efficient, community-built agent—optimizing for quality and speed while rewarding the creators behind the best-performing agents. This was a defining moment in realizing a truly community-owned intelligence layer. ✅- From agents to a living agent network: With the launch of the Agent Social Network and Super Agent architecture, ChainOpera AI moved beyond isolated agents toward a collaborative system where humans and specialized agents coordinate, share context, and solve complex, multi-step tasks together. ✅- $COAI breakout year: The listing of $COAI across major exchanges shocked the market, and throughout the year COAI consistently remained among the top AI-native crypto tokens by visibility, activity, and community engagement – reflecting growing confidence in the long-term vision of collaborative intelligence. ✅- Global presence: ChainOpera AI around-the-world tour: ChainOpera AI went global in 2025, sponsoring and participating in major AI and Web3 events across North America, Europe, and Asia, including ETHDenver, Consensus Toronto, Token2049 Singapore, ETHCC, SBC, and Devcon. These global touchpoints helped us engage directly with developers, builders, investors, and partners worldwide, accelerating adoption and positioning ChainOpera AI at the center of the emerging AIxBlockchain movement. ✅- Community momentum at scale: Community remained the heart of ChainOpera AI’s growth. We successfully completed three seasons of structured community engagement, executed a widely participated community airdrop, and ran multiple ecosystem-shaping campaigns to incentivize builders, creators, and early adopters. These efforts strengthened alignment between users, developers, and the protocol, laying the foundation for a durable, community-owned AI ecosystem. ✅- “AI for Markets” taking shape: We laid critical groundwork for AI-native market intelligence, including the launch of PrediMarket Agent and multiple trading and analysis agents—early building blocks toward an AI-driven ecosystem for crypto and DeFi markets. ✅- Building in public, with the community: Across product launches, research milestones, ecosystem discussions, and global events, we continued to build openly to bring developers, users, and partners directly into the evolution of ChainOpera AI. This year also marked the launch of the ChainOpera AI Foundation website, formally kicking off a bold Ecosystem Fund designed to empower builders, incubate high-impact projects, and accelerate the growth of a truly community-owned, collaborative AI ecosystem. To every builder, user, and supporter who helped make this year possible: THANK YOU! 🧭 What we’re excited about in the coming year 🔹- A Stronger, Denser Agent Economy (everyday adoption + cross-chain reach): In 2026, we are scaling the Agent Economy from growth to daily usage, with more agents, richer workflows, deeper multi-agent collaboration, and higher-impact use cases that users rely on every day. In parallel, we are expanding the agent network beyond a single ecosystem with cross-chain execution and interoperability, allowing agents to access the best liquidity, data, and opportunities wherever they exist. 🔹- AI Market Infrastructure Evolution: Building on PrediMarket Agent and our growing suite of trading and market-intelligence agents, we are advancing toward a mature AI market infrastructure, where agents continuously monitor, reason, simulate, optimize, and act across crypto, DeFi, and beyond. The goal is to make complex markets more accessible, more transparent, and more intelligence-driven, turning research, decision-making, and execution into a fast and reliable loop for everyday users. 🔹- Ecosystem Acceleration through the Foundation: With the ChainOpera AI Foundation and our Ecosystem Fund and Co-Creation Grants, we are doubling down on empowering independent builders to expand the protocol, the agent network, and the underlying infrastructure, so the community can co-create, co-own, and scale the ecosystem together. 🔹- Business Expansion and Market Penetration: In 2026, we will focus on expanding ChainOpera’s reach through strategic partnerships, product-led growth, and new paths to monetization, bringing AI agents to a broader global user base and driving sustained adoption, engagement, and revenue, while staying aligned with community ownership and an open ecosystem. 2025 was the proof. 2026 is where it compounds. 🔥 Co-Create. Co-Own. COAI.
0:22

What a year. 🚀 2025 was the year ChainOpera AI turned vision into real momentum: building a community-co-created, community-co-owned AI agent network and pushing the boundaries of what decentralized, collaborative intelligence can look like. 🚀 Biggest highlights from 2025 ✅- AI Terminal officially launched: We unveiled the ChainOpera AI Terminal as a unified gateway to decentralized AI, making it possible for anyone to interact with powerful, decentralized LLMs without technical friction. Positioned as the “browser for the DeAI era,” the AI Terminal marked a major step toward making decentralized intelligence accessible, usable, and mainstream. ✅- AI Terminal adoption at massive scale: Momentum followed quickly. The AI Terminal surpassed 2M registered users and consistently ranked top 3 among all apps on the BNB AI DappBay, validating strong product–market fit and real, sustained usage at scale. ✅- Announcing Coco: the world’s first community-owned Super Agent: We introduced Coco, the intelligence layer that sits between users and the agent network. Coco dynamically routes each request to the most efficient, community-built agent—optimizing for quality and speed while rewarding the creators behind the best-performing agents. This was a defining moment in realizing a truly community-owned intelligence layer. ✅- From agents to a living agent network: With the launch of the Agent Social Network and Super Agent architecture, ChainOpera AI moved beyond isolated agents toward a collaborative system where humans and specialized agents coordinate, share context, and solve complex, multi-step tasks together. ✅- $COAI breakout year: The listing of $COAI across major exchanges shocked the market, and throughout the year COAI consistently remained among the top AI-native crypto tokens by visibility, activity, and community engagement – reflecting growing confidence in the long-term vision of collaborative intelligence. ✅- Global presence: ChainOpera AI around-the-world tour: ChainOpera AI went global in 2025, sponsoring and participating in major AI and Web3 events across North America, Europe, and Asia, including ETHDenver, Consensus Toronto, Token2049 Singapore, ETHCC, SBC, and Devcon. These global touchpoints helped us engage directly with developers, builders, investors, and partners worldwide, accelerating adoption and positioning ChainOpera AI at the center of the emerging AIxBlockchain movement. ✅- Community momentum at scale: Community remained the heart of ChainOpera AI’s growth. We successfully completed three seasons of structured community engagement, executed a widely participated community airdrop, and ran multiple ecosystem-shaping campaigns to incentivize builders, creators, and early adopters. These efforts strengthened alignment between users, developers, and the protocol, laying the foundation for a durable, community-owned AI ecosystem. ✅- “AI for Markets” taking shape: We laid critical groundwork for AI-native market intelligence, including the launch of PrediMarket Agent and multiple trading and analysis agents—early building blocks toward an AI-driven ecosystem for crypto and DeFi markets. ✅- Building in public, with the community: Across product launches, research milestones, ecosystem discussions, and global events, we continued to build openly to bring developers, users, and partners directly into the evolution of ChainOpera AI. This year also marked the launch of the ChainOpera AI Foundation website, formally kicking off a bold Ecosystem Fund designed to empower builders, incubate high-impact projects, and accelerate the growth of a truly community-owned, collaborative AI ecosystem. To every builder, user, and supporter who helped make this year possible: THANK YOU! 🧭 What we’re excited about in the coming year 🔹- A Stronger, Denser Agent Economy (everyday adoption + cross-chain reach): In 2026, we are scaling the Agent Economy from growth to daily usage, with more agents, richer workflows, deeper multi-agent collaboration, and higher-impact use cases that users rely on every day. In parallel, we are expanding the agent network beyond a single ecosystem with cross-chain execution and interoperability, allowing agents to access the best liquidity, data, and opportunities wherever they exist. 🔹- AI Market Infrastructure Evolution: Building on PrediMarket Agent and our growing suite of trading and market-intelligence agents, we are advancing toward a mature AI market infrastructure, where agents continuously monitor, reason, simulate, optimize, and act across crypto, DeFi, and beyond. The goal is to make complex markets more accessible, more transparent, and more intelligence-driven, turning research, decision-making, and execution into a fast and reliable loop for everyday users. 🔹- Ecosystem Acceleration through the Foundation: With the ChainOpera AI Foundation and our Ecosystem Fund and Co-Creation Grants, we are doubling down on empowering independent builders to expand the protocol, the agent network, and the underlying infrastructure, so the community can co-create, co-own, and scale the ecosystem together. 🔹- Business Expansion and Market Penetration: In 2026, we will focus on expanding ChainOpera’s reach through strategic partnerships, product-led growth, and new paths to monetization, bringing AI agents to a broader global user base and driving sustained adoption, engagement, and revenue, while staying aligned with community ownership and an open ecosystem. 2025 was the proof. 2026 is where it compounds. 🔥 Co-Create. Co-Own. COAI.

ChainOpera AI

17,042 views • 7 months ago

$AMD $NVDA & the AMD Bear SemiAnalysis 🧵 Here are some facts: $META allocated 42% AI GPUs to $AMD OpenAI allocated 6GW(38%) to $AMD 1. Model-Specific Bias: Llama 3.3 70B graph favored NVIDIA due to TRT-LLM optimizations, highlighting throughput and latency where Blackwell excels. In contrast, the GPT-OSS 120B chart shifts focus to cost and interactivity, where MI355X shines. This selective model choice clearly suggests SemiAnalysis tailors benchmarks to reinforce narratives—NVIDIA’s dominance in speed (Llama 3.3) and AMD’s niche in cost (GPT-OSS). GPT-OSS 120B, with its sparse attention mechanisms (similar to DeepSeek-V3.2-Exp), shows AMD’s CDNA 4 architecture, while Llama 3.3’s dense attention favors NVIDIA’s Tensor Cores. SemiAnalysis’ decision to emphasize Llama 3.3 initially could reflect its AMD bear stance. 2. The way Data is presented The Llama 3.3 graph focused on raw performance metrics (throughput vs. latency), downplaying cost, where AMD holds an edge. This new chart, buried in follow-up posts, reveals AMD’s strength but receives less prominence, suggesting a curated narrative. Labeling variability (e.g., B200 with/without TRT) and the lack of uniform scaling across graphs indicate potential cherry-picking of configurations to favor NVIDIA’s optimized setups. 3. Historical Context: SemiAnalysis’ past critiques of AMD’s R&D and ROCm (web results from May 2025) align with a bearish outlook. Their own hype/brand around NVIDIA’s 15x ROI contrasts with muted coverage of AMD’s cost advantages, reinforcing bias. Despite AMD’s participation in InferenceMAX, the benchmark’s framing (e.g., prioritizing Blackwell’s ROI) reflect SemiAnalysis’ market predictions rather than balanced analysis. Lastly, AMD’s Instinct MI355X proves superior in inference and cost per million tokens for the GPT-OSS 120B model, offering a 25% cost advantage over NVIDIA’s H200 at moderate-to-high interactivity levels. This efficiency, driven by AMD’s memory bandwidth and FP4 support, makes it a better choice for cost-sensitive, multi-user deployments over a three-year horizon. However, SemiAnalysis’ sole focus(presentation graph) on Llama 3.3—where NVIDIA excels demonstrates a pattern of cherry-picking models and data to favor NVIDIA , consistent with its historical AMD bearish stance. This selective presentation risks misleading stakeholders by overshadowing AMD economic strengths. My personal take: I would trust Dr. Lisa Su, and Greg Brockman Sam Altman take on AMD and how they viewed and allocated 6GW for AMD over SemiAnalysis . At the end of the day, Large customers pay when it works. $Meta allocated 42% AI GPUs to $AMD for a reason. And the "secret weapon" will improve energy consumption by 20-50%, meaning at 6GW, OpenAI would be able to deploy 25-50% more MI450 at a much better cost advantage, higher memory bandwidth, and the queen of Inference! Oh and ROCm 8 is expected to be on par with CUDA in 2026.
2:46

$AMD $NVDA & the AMD Bear SemiAnalysis 🧵 Here are some facts: $META allocated 42% AI GPUs to $AMD OpenAI allocated 6GW(38%) to $AMD 1. Model-Specific Bias: Llama 3.3 70B graph favored NVIDIA due to TRT-LLM optimizations, highlighting throughput and latency where Blackwell excels. In contrast, the GPT-OSS 120B chart shifts focus to cost and interactivity, where MI355X shines. This selective model choice clearly suggests SemiAnalysis tailors benchmarks to reinforce narratives—NVIDIA’s dominance in speed (Llama 3.3) and AMD’s niche in cost (GPT-OSS). GPT-OSS 120B, with its sparse attention mechanisms (similar to DeepSeek-V3.2-Exp), shows AMD’s CDNA 4 architecture, while Llama 3.3’s dense attention favors NVIDIA’s Tensor Cores. SemiAnalysis’ decision to emphasize Llama 3.3 initially could reflect its AMD bear stance. 2. The way Data is presented The Llama 3.3 graph focused on raw performance metrics (throughput vs. latency), downplaying cost, where AMD holds an edge. This new chart, buried in follow-up posts, reveals AMD’s strength but receives less prominence, suggesting a curated narrative. Labeling variability (e.g., B200 with/without TRT) and the lack of uniform scaling across graphs indicate potential cherry-picking of configurations to favor NVIDIA’s optimized setups. 3. Historical Context: SemiAnalysis’ past critiques of AMD’s R&D and ROCm (web results from May 2025) align with a bearish outlook. Their own hype/brand around NVIDIA’s 15x ROI contrasts with muted coverage of AMD’s cost advantages, reinforcing bias. Despite AMD’s participation in InferenceMAX, the benchmark’s framing (e.g., prioritizing Blackwell’s ROI) reflect SemiAnalysis’ market predictions rather than balanced analysis. Lastly, AMD’s Instinct MI355X proves superior in inference and cost per million tokens for the GPT-OSS 120B model, offering a 25% cost advantage over NVIDIA’s H200 at moderate-to-high interactivity levels. This efficiency, driven by AMD’s memory bandwidth and FP4 support, makes it a better choice for cost-sensitive, multi-user deployments over a three-year horizon. However, SemiAnalysis’ sole focus(presentation graph) on Llama 3.3—where NVIDIA excels demonstrates a pattern of cherry-picking models and data to favor NVIDIA , consistent with its historical AMD bearish stance. This selective presentation risks misleading stakeholders by overshadowing AMD economic strengths. My personal take: I would trust Dr. Lisa Su, and Greg Brockman Sam Altman take on AMD and how they viewed and allocated 6GW for AMD over SemiAnalysis . At the end of the day, Large customers pay when it works. $Meta allocated 42% AI GPUs to $AMD for a reason. And the "secret weapon" will improve energy consumption by 20-50%, meaning at 6GW, OpenAI would be able to deploy 25-50% more MI450 at a much better cost advantage, higher memory bandwidth, and the queen of Inference! Oh and ROCm 8 is expected to be on par with CUDA in 2026.

Mike

104,194 views • 10 months ago