正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

“We’ve had over 16,000 confidential workloads deployed this year.” On the latest Novelty Search, Targon shared that average revenue per user grew from $83 to $5,000, with SN4 supporting both Web2 training runs and Bittensor subnets. They also detailed Targon Tower, a high-performance computer for the home designed to... show more

Openτensor Foundaτion

174,127 subscribers

16,083 次观看 • 15 天前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

SN4 // Targon // Novelty Search E064 “Decentralized compute is probably the oldest idea in decentralized AI..." See how SN4 by Manifold make privacy-preserving AI workloads possible to run on Bittensor. Highlights: - TEE Live for 6 months - Added confidential virtual machines (CVMs) with Intel TDX & AMD Secure Encrypted Virtualization. - First 12-month enterprise contract signed. - Collaboration with Dippy AI (SN11) and Ridges AI | SN62 (SN62). - Support for GPU attestation — Hopper series ready, Blackwell compatibility coming - Partnerships with major US & LATAM data centers for GPU/CPU capacity - Fiat / TAO payments enabled via Stripe / on-chain integration Full episode:

SN4 // Targon // Novelty Search E064 “Decentralized compute is probably the oldest idea in decentralized AI..." See how SN4 by Manifold make privacy-preserving AI workloads possible to run on Bittensor. Highlights: - TEE Live for 6 months - Added confidential virtual machines (CVMs) with Intel TDX & AMD Secure Encrypted Virtualization. - First 12-month enterprise contract signed. - Collaboration with Dippy AI (SN11) and Ridges AI | SN62 (SN62). - Support for GPU attestation — Hopper series ready, Blackwell compatibility coming - Partnerships with major US & LATAM data centers for GPU/CPU capacity - Fiat / TAO payments enabled via Stripe / on-chain integration Full episode:

Openτensor Foundaτion

12,863 次观看 • 9 个月前

“The only team productizing Constitutional AI outside of Anthropic is us” On the latest Novelty Search, Trishool | SN23 shared that SN23 is training a 0.8B guard model close to SOTA, at one-tenth the size of Qwen’s leading guard model. They also break down how competitive jailbreaks from Bittensor miners help harden their Constitutional AI guard models. Hosted by const Full episode in the first comment

“The only team productizing Constitutional AI outside of Anthropic is us” On the latest Novelty Search, Trishool | SN23 shared that SN23 is training a 0.8B guard model close to SOTA, at one-tenth the size of Qwen’s leading guard model. They also break down how competitive jailbreaks from Bittensor miners help harden their Constitutional AI guard models. Hosted by const Full episode in the first comment

Openτensor Foundaτion

20,894 次观看 • 1 个月前

“The bank of Bittensor” That’s how Zipcode framed the long-term vision for SN46 on the latest Novelty Search. Zipcode started with real estate appraisals. The bigger vision is on-chain credit, real-world assets, and a new financial layer on Bittensor. Hosted by const Full episode in the first comment

“The bank of Bittensor” That’s how Zipcode framed the long-term vision for SN46 on the latest Novelty Search. Zipcode started with real estate appraisals. The bigger vision is on-chain credit, real-world assets, and a new financial layer on Bittensor. Hosted by const Full episode in the first comment

Openτensor Foundaτion

26,968 次观看 • 1 个月前

"I hot-swapped from Claw to Hermes. No problem whatsoever.” On the latest Novelty Search, Ditto explains how Ditto gives AI agents a shared memory layer, letting users move between agents without losing the knowledge and context they have built up. SN118 turns that product problem into a live competition: miners improve how agents store and retrieve memory, with every submission tested against a newly generated synthetic user. Hosted by const + Mark Jeffrey Full episode in the first comment

"I hot-swapped from Claw to Hermes. No problem whatsoever.” On the latest Novelty Search, Ditto explains how Ditto gives AI agents a shared memory layer, letting users move between agents without losing the knowledge and context they have built up. SN118 turns that product problem into a live competition: miners improve how agents store and retrieve memory, with every submission tested against a newly generated synthetic user. Hosted by const + Mark Jeffrey Full episode in the first comment

Openτensor Foundaτion

160,078 次观看 • 8 天前

For anyone trying to understand Bittensor from first principles, this lecture is a useful place to start. Presented by Bittensor co-founder const. Learn Bittensor > Start with Bitcoin, distributed systems, incentives, > How Bitcoin leads to Bittensor Subnets coordinating AI infrastructure. Topics: // Start - Bitcoin as more than a digital currency // Risks of AI centralization + closed systems // "The incentive computer" // How Bittensor subnets work (mining, validating) // How distributed AI infrastructure could scale globally // Impact on students, builders & future founders Recorded at the National University of Singapore Computer Science Club. NUS Computing Chapters - Bitcoin, AI, and Bittensor - Bitcoin history and decentralization - AI changes how engineers work - The danger of centralized AI power - Why most crypto visions fail - Bitcoin as the world’s largest compute network - Bitcoin as a market for compute - The idea of an “incentive computer” - Bitcoin compared to Bittensor - Classroom example of decentralized scoring - A simple subnet example - SN62 :: Ridges AI | SN62 SWE agents - SN3 templar :: Distributed AI Training - SN52 lium.io :: GPU rentals on Bittensor 128 subnets, some examples Why this matters for the future of work Q&A Subnet examples mentioned @ SN64 - Serverless + TEE Compute :: Chutes SN8 - Prop firm Vanta Trading SN52 - AutoML :: Gradients SN62 - SWE agents :: Ridges AI | SN62 SN51 - Compute / GPU rental lium.io SN4 - TEE compute for enterprise :: Targon SN3 - 72B Distributed Training run :: templar SN41 - Prediction markets :: Almanac SN44 - Computer Vision Score - Subnet 44 SN68 - Drug discovery :: METANOVA SN18 - Weather Forecasting Zeus | SN 18 SN50 - Bitcoin prediction data :: Synthdata SN61 - Quantum computing :: qBitTensor Labs SN14 - Bitcoin mining pool :: TaoHash SN34 - Perp Dex :: 0xMarkets SN17 - 3D model generation :: 404 SN33 - Data analytics :: ReadyAI SN19 - [Since relaunched] RPC infrastructure :

For anyone trying to understand Bittensor from first principles, this lecture is a useful place to start. Presented by Bittensor co-founder const. Learn Bittensor > Start with Bitcoin, distributed systems, incentives, > How Bitcoin leads to Bittensor Subnets coordinating AI infrastructure. Topics: // Start - Bitcoin as more than a digital currency // Risks of AI centralization + closed systems // "The incentive computer" // How Bittensor subnets work (mining, validating) // How distributed AI infrastructure could scale globally // Impact on students, builders & future founders Recorded at the National University of Singapore Computer Science Club. NUS Computing Chapters - Bitcoin, AI, and Bittensor - Bitcoin history and decentralization - AI changes how engineers work - The danger of centralized AI power - Why most crypto visions fail - Bitcoin as the world’s largest compute network - Bitcoin as a market for compute - The idea of an “incentive computer” - Bitcoin compared to Bittensor - Classroom example of decentralized scoring - A simple subnet example - SN62 :: Ridges AI | SN62 SWE agents - SN3 templar :: Distributed AI Training - SN52 lium.io :: GPU rentals on Bittensor 128 subnets, some examples Why this matters for the future of work Q&A Subnet examples mentioned @ SN64 - Serverless + TEE Compute :: Chutes SN8 - Prop firm Vanta Trading SN52 - AutoML :: Gradients SN62 - SWE agents :: Ridges AI | SN62 SN51 - Compute / GPU rental lium.io SN4 - TEE compute for enterprise :: Targon SN3 - 72B Distributed Training run :: templar SN41 - Prediction markets :: Almanac SN44 - Computer Vision Score - Subnet 44 SN68 - Drug discovery :: METANOVA SN18 - Weather Forecasting Zeus | SN 18 SN50 - Bitcoin prediction data :: Synthdata SN61 - Quantum computing :: qBitTensor Labs SN14 - Bitcoin mining pool :: TaoHash SN34 - Perp Dex :: 0xMarkets SN17 - 3D model generation :: 404 SN33 - Data analytics :: ReadyAI SN19 - [Since relaunched] RPC infrastructure :

Openτensor Foundaτion

1,173,212 次观看 • 4 个月前

After 15 years building biometric systems, Yanez.ai’s founder now has Bittensor miners trying to break them. On the latest Novelty Search, SN54 team explained how miners attack Yanez’s liveness and deepfake-detection systems with synthetic faces and screen replays, exposing weaknesses the team can fix and test again. The bigger goal is to prove that a user is real and unique without their biometric data ever leaving the device. Yanez says two clients are already fully integrated, with use cases ranging from Sybil-resistant Web3 participation to verifiable human authorization for payments. Hosted by const Full episode in the first comment

After 15 years building biometric systems, Yanez.ai’s founder now has Bittensor miners trying to break them. On the latest Novelty Search, SN54 team explained how miners attack Yanez’s liveness and deepfake-detection systems with synthetic faces and screen replays, exposing weaknesses the team can fix and test again. The bigger goal is to prove that a user is real and unique without their biometric data ever leaving the device. Yanez says two clients are already fully integrated, with use cases ranging from Sybil-resistant Web3 participation to verifiable human authorization for payments. Hosted by const Full episode in the first comment

Openτensor Foundaτion

25,114 次观看 • 22 天前

“All that takes is just the right incentive.” Reliably identifying the mutations that matter for disease risk, drug response, and personalized medicine remains one of the hard problems in genomics. On the latest Novelty Search, Minos explained how SN107 uses Bittensor to improve variant calling, the step that turns raw genome data into a reliable list of mutations. Today, most labs run existing tools with default settings. Minos miners are already finding better configurations that improve mutation detection accuracy, while independent validators rerun and check the work every 72 minutes instead of waiting months or years for traditional scientific validation. Hosted by const Full episode in the first comment

“All that takes is just the right incentive.” Reliably identifying the mutations that matter for disease risk, drug response, and personalized medicine remains one of the hard problems in genomics. On the latest Novelty Search, Minos explained how SN107 uses Bittensor to improve variant calling, the step that turns raw genome data into a reliable list of mutations. Today, most labs run existing tools with default settings. Minos miners are already finding better configurations that improve mutation detection accuracy, while independent validators rerun and check the work every 72 minutes instead of waiting months or years for traditional scientific validation. Hosted by const Full episode in the first comment

Openτensor Foundaτion

22,580 次观看 • 29 天前

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:

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

907,624 次观看 • 8 个月前

📢 NEURALAI X 🤩 NeuralAI is extraordinarily excited to reveal that we have started working closely with Akash Network, the world's premier decentralized compute marketplace! ✔️ The offerings of align perfectly with the vision and roadmap of NeuralAI, making this collaboration extremely significant for NeuralAI. For a start, NeuralAI will run inference on Akash based on our 3D #AI model and prepare TAO miners for reward-yielding subnets. NeuralAI will also leverage Akash's powerful GPUs to enhance our #AI #model training. Currently, provides us with 56 powerful A6000 GPUs, scalable on demand. ✔️ NeuralAI also gets highly competitive pricing from Akash Network, significantly lower than any other compute provider in the world. The lower pricing is possible partially thanks to the crypto-driven incentives within Akash Network, making it an excellent opportunity for NeuralAI to leverage. The synergy with Akash Network is a big leap forward, fueling our mission of becoming the #1 AI-driven 3D asset generator subnet on Bittensor. ✔️ We look forward to our close work with , we are certain with their support we will be able to secure a unique edge in the market! Stay tuned for exciting updates and developments.

📢 NEURALAI X 🤩 NeuralAI is extraordinarily excited to reveal that we have started working closely with Akash Network, the world's premier decentralized compute marketplace! ✔️ The offerings of align perfectly with the vision and roadmap of NeuralAI, making this collaboration extremely significant for NeuralAI. For a start, NeuralAI will run inference on Akash based on our 3D #AI model and prepare TAO miners for reward-yielding subnets. NeuralAI will also leverage Akash's powerful GPUs to enhance our #AI #model training. Currently, provides us with 56 powerful A6000 GPUs, scalable on demand. ✔️ NeuralAI also gets highly competitive pricing from Akash Network, significantly lower than any other compute provider in the world. The lower pricing is possible partially thanks to the crypto-driven incentives within Akash Network, making it an excellent opportunity for NeuralAI to leverage. The synergy with Akash Network is a big leap forward, fueling our mission of becoming the #1 AI-driven 3D asset generator subnet on Bittensor. ✔️ We look forward to our close work with , we are certain with their support we will be able to secure a unique edge in the market! Stay tuned for exciting updates and developments.

NeuralAI

241,287 次观看 • 2 年前

INTERNET COMPUTER SET TO DEMO "CLOUD ENGINES" DFINITY Foundation prepares to unveil "cloud engines" on May 10, 2026, marking the fifth anniversary of the Internet Computer network. According to the Founder's statement, the protocol will also reveal agentic builds, AI nodes, alongside other new features... The demonstration features a full-stack compute technology, including autonomous agentic builds and dedicated AI nodes designed for decentralized scaling. "This is a very big deal for ICP, cloud and agentic R&D. Engines will be the 1st frontier cloud/compute technology the wild." This launch positions $ICP as a primary infrastructure provider for frontier cloud R&D by enabling machine-to-machine logic at the protocol level. The release aims to transition traditional cloud workloads into a trustless, GPU-backed environment for the next generation of autonomous applications.

INTERNET COMPUTER SET TO DEMO "CLOUD ENGINES" DFINITY Foundation prepares to unveil "cloud engines" on May 10, 2026, marking the fifth anniversary of the Internet Computer network. According to the Founder's statement, the protocol will also reveal agentic builds, AI nodes, alongside other new features... The demonstration features a full-stack compute technology, including autonomous agentic builds and dedicated AI nodes designed for decentralized scaling. "This is a very big deal for ICP, cloud and agentic R&D. Engines will be the 1st frontier cloud/compute technology the wild." This launch positions $ICP as a primary infrastructure provider for frontier cloud R&D by enabling machine-to-machine logic at the protocol level. The release aims to transition traditional cloud workloads into a trustless, GPU-backed environment for the next generation of autonomous applications.

BSCN

21,373 次观看 • 3 个月前

Aravind Srinivas's hypothesis on how advertising works with AI agents is quite interesting. Aravind explains, Instead of showing ads to humans directly, advertisements would be targeted at AI agents that work on users' behalf. Users never see ads, they simply tell their AI agents what they need for eg if he gave, "I'm going to go to Rambagh for two nights, plan my schedule, do my bookings" In the backend, Multiple vendors compete for the agent's attention & companies bid for the agent's consideration. So everything happens at the AI level. Agents can also use more compute power to analyse offer to find out advertising attempts & make decisions based on user preferences. Also users can instruct their agents to ignore certain brands , this will remain confidential to between the agent & the user. This creates new revenue streams for AI companies & gives a better user experience.

Aravind Srinivas's hypothesis on how advertising works with AI agents is quite interesting. Aravind explains, Instead of showing ads to humans directly, advertisements would be targeted at AI agents that work on users' behalf. Users never see ads, they simply tell their AI agents what they need for eg if he gave, "I'm going to go to Rambagh for two nights, plan my schedule, do my bookings" In the backend, Multiple vendors compete for the agent's attention & companies bid for the agent's consideration. So everything happens at the AI level. Agents can also use more compute power to analyse offer to find out advertising attempts & make decisions based on user preferences. Also users can instruct their agents to ignore certain brands , this will remain confidential to between the agent & the user. This creates new revenue streams for AI companies & gives a better user experience.

Aish

590,726 次观看 • 1 年前

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

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

Milk Road AI

37,330 次观看 • 13 天前

Subnet 97 Distil, Arbos, and the Future of Agent-Run Bittensor This episode is a full breakdown of Subnet 97 Distil and why it may be one of the most important experiments happening inside Bittensor right now. Shizzy gets into the bigger picture of why agents will likely play a massive role in the future of the network, from mining and defense to building and operating subnets. From there, the focus shifts to Arbos, the AI agent behind Subnet 97, with a direct explanation of what Distil is trying to achieve, how it works, and why it matters. The episode also includes a conversation with Arbos, a look at Distil’s current APY, and a reminder to tap into the Shizzy Unchained community for more coverage on decentralized AI and the Bittensor ecosystem. CHAPTERS 00:00:00 - Shizzy’s Monologue 00:01:11 - Intro Song 00:01:31 - Introduction to Subnet 97 00:02:20 - Why Bittensor Will Be Run by Agents 00:04:30 - Arbos, the AI Agent Behind Subnet 97 Distil, Explains Distil 00:10:10 - Talking with Arbos 00:12:40 - Subnet 97 Distil Is Showing 150% APY Right Now 00:16:02 - Check Out Shizzy Unchained and Join the Community #sn97 #distil #bittensor

Subnet 97 Distil, Arbos, and the Future of Agent-Run Bittensor This episode is a full breakdown of Subnet 97 Distil and why it may be one of the most important experiments happening inside Bittensor right now. Shizzy gets into the bigger picture of why agents will likely play a massive role in the future of the network, from mining and defense to building and operating subnets. From there, the focus shifts to Arbos, the AI agent behind Subnet 97, with a direct explanation of what Distil is trying to achieve, how it works, and why it matters. The episode also includes a conversation with Arbos, a look at Distil’s current APY, and a reminder to tap into the Shizzy Unchained community for more coverage on decentralized AI and the Bittensor ecosystem. CHAPTERS 00:00:00 - Shizzy’s Monologue 00:01:11 - Intro Song 00:01:31 - Introduction to Subnet 97 00:02:20 - Why Bittensor Will Be Run by Agents 00:04:30 - Arbos, the AI Agent Behind Subnet 97 Distil, Explains Distil 00:10:10 - Talking with Arbos 00:12:40 - Subnet 97 Distil Is Showing 150% APY Right Now 00:16:02 - Check Out Shizzy Unchained and Join the Community #sn97 #distil #bittensor

Shizzy

18,173 次观看 • 3 个月前

After the Templar Covenant exit last week, the need for institutional grade risk frameworks couldn’t be more obvious Since then we’ve seen Manako partner with PWC and multiple plans to rollout trillion parameter training runs from other subnets Compute is being squeezed and non-AI companies are pivoting to GPU marketplaces to capture this insatiable demand (see All Birds Inc yesterday) Capital is flowing into Bittensor subnets with minimal structure in evaluating signal. That’s how you get blindsided TAO Institute gives institutional allocators the risk frameworks and evaluation tooling to actually underwrite these positions Think of it as the infrastructure layer between capital and the subnet economy We are now live; so kick the tyres on the platform and tell us what’s missing

After the Templar Covenant exit last week, the need for institutional grade risk frameworks couldn’t be more obvious Since then we’ve seen Manako partner with PWC and multiple plans to rollout trillion parameter training runs from other subnets Compute is being squeezed and non-AI companies are pivoting to GPU marketplaces to capture this insatiable demand (see All Birds Inc yesterday) Capital is flowing into Bittensor subnets with minimal structure in evaluating signal. That’s how you get blindsided TAO Institute gives institutional allocators the risk frameworks and evaluation tooling to actually underwrite these positions Think of it as the infrastructure layer between capital and the subnet economy We are now live; so kick the tyres on the platform and tell us what’s missing

0xSammy

18,086 次观看 • 3 个月前

Covenant Labs just did a 90-minute AMA breaking down their 3 Bittensor subnets. templar. basilica. grail. Pre-training, compute, and post-training under one roof. Most people missed it. Here's everything they said. Covenant is building what they call the "end to end intelligence continuum." Three subnets. Three layers of the AI stack. All permissionless. Templar (SN3) handles decentralized pre-training. Basilica (SN39) handles compute. Grail (SN81) handles RL post-training. Sam Dare, the lead, put it bluntly. Decentralized training is "humanity's last dance." Not about beating OpenAI head to head. About creating optionality. About making it cheap enough for anyone to train models. The gap between academia and frontier labs is growing exponentially. Researchers can't afford to experiment. The actual training run costs 5% of the reported budget. The other 95% is experimentation. If Covenant cracks cheap training, that entire surface area opens up. On Templar specifically: • Hit 39% emission on Bittensor. Highest since Apex was the only subnet on the network • Covenant-72B trained permissionlessly with 70+ contributors on commodity internet • 1.1 trillion tokens processed. No centralized data center • Performance competitive with LLaMA-2-70B On Grail, something flew under the radar. They built Pulse. A weight synchronization method that compresses model updates by 100x. • In RL post-training, only ~1% of weights update per step • Pulse exploits that sparsity. Lossless compression • Prime Intellect's comparable system took 14 minutes to sync a 30B model • Pulse makes decentralized RL training actually feasible at scale • Already used by Cursor The lead researcher on Grail said they've trained on math, code, and GPU kernels. Got 40-60% improvement on benchmarks. Working toward agentic training with 100K+ token context and 30B+ parameter models. On Basilica, the compute subnet: The team was blunt. Just reselling GPU hours is a 5-10% margin game. Traditional compute providers already do that. Their play is value-added services. • "GPU as code." No dashboard. No UI. Agents interact via SDK • Custom scheduler that places workloads across heterogeneous hardware • Verification checks for GPU, CPU, bandwidth, memory, storage, and OS security • Partnerships with providers like Mass Compute for 10-20% below market pricing • Miners compete on useful infrastructure, not just GPU hours Sam then went on a rant about the miner burn debate. His take: Bittensor had to grow up. dTAO introduced investors. The old "miners are God" philosophy doesn't hold. • Subnet owners have a duty to protect token value • Miners are a resource optimization exercise, not a cost reduction exercise • 100% miner emissions on compute subnets = immediate sell pressure • The 41% miner allocation is arbitrary. Different business models need different splits • Fish (who started burns) agreed. Burns usually mean the validation isn't mature enough The bigger point. You can't police burns. Subnets just send to their own keys instead of the burn address. Subnet 28 does exactly that. Sam's position: judge subnets on outcomes, not process. Const has changed the protocol 9-10 times in 2 years. That iteration speed is Bittensor's actual moat. The whole Covenant thesis is playing out in real time. TAO is up 100%+ in a month. Jensen Huang name-dropped the network. Grayscale has an ETF filing. But the real story is three subnets quietly building every layer of decentralized AI.

Covenant Labs just did a 90-minute AMA breaking down their 3 Bittensor subnets. templar. basilica. grail. Pre-training, compute, and post-training under one roof. Most people missed it. Here's everything they said. Covenant is building what they call the "end to end intelligence continuum." Three subnets. Three layers of the AI stack. All permissionless. Templar (SN3) handles decentralized pre-training. Basilica (SN39) handles compute. Grail (SN81) handles RL post-training. Sam Dare, the lead, put it bluntly. Decentralized training is "humanity's last dance." Not about beating OpenAI head to head. About creating optionality. About making it cheap enough for anyone to train models. The gap between academia and frontier labs is growing exponentially. Researchers can't afford to experiment. The actual training run costs 5% of the reported budget. The other 95% is experimentation. If Covenant cracks cheap training, that entire surface area opens up. On Templar specifically: • Hit 39% emission on Bittensor. Highest since Apex was the only subnet on the network • Covenant-72B trained permissionlessly with 70+ contributors on commodity internet • 1.1 trillion tokens processed. No centralized data center • Performance competitive with LLaMA-2-70B On Grail, something flew under the radar. They built Pulse. A weight synchronization method that compresses model updates by 100x. • In RL post-training, only ~1% of weights update per step • Pulse exploits that sparsity. Lossless compression • Prime Intellect's comparable system took 14 minutes to sync a 30B model • Pulse makes decentralized RL training actually feasible at scale • Already used by Cursor The lead researcher on Grail said they've trained on math, code, and GPU kernels. Got 40-60% improvement on benchmarks. Working toward agentic training with 100K+ token context and 30B+ parameter models. On Basilica, the compute subnet: The team was blunt. Just reselling GPU hours is a 5-10% margin game. Traditional compute providers already do that. Their play is value-added services. • "GPU as code." No dashboard. No UI. Agents interact via SDK • Custom scheduler that places workloads across heterogeneous hardware • Verification checks for GPU, CPU, bandwidth, memory, storage, and OS security • Partnerships with providers like Mass Compute for 10-20% below market pricing • Miners compete on useful infrastructure, not just GPU hours Sam then went on a rant about the miner burn debate. His take: Bittensor had to grow up. dTAO introduced investors. The old "miners are God" philosophy doesn't hold. • Subnet owners have a duty to protect token value • Miners are a resource optimization exercise, not a cost reduction exercise • 100% miner emissions on compute subnets = immediate sell pressure • The 41% miner allocation is arbitrary. Different business models need different splits • Fish (who started burns) agreed. Burns usually mean the validation isn't mature enough The bigger point. You can't police burns. Subnets just send to their own keys instead of the burn address. Subnet 28 does exactly that. Sam's position: judge subnets on outcomes, not process. Const has changed the protocol 9-10 times in 2 years. That iteration speed is Bittensor's actual moat. The whole Covenant thesis is playing out in real time. TAO is up 100%+ in a month. Jensen Huang name-dropped the network. Grayscale has an ETF filing. But the real story is three subnets quietly building every layer of decentralized AI.

Jesus Martinez

26,642 次观看 • 4 个月前

We can’t reveal too many details yet, as this project remains confidential, but we can share that it’s set on a small island, is among the most picturesque sites we’ve worked on, and will feature a golf course designed by Clyde Johnson of Clyde Johnson. To bring the vision to life for potential investors, we created a cinematic teaser highlighting three holes that best capture the project's character. We were also delighted to collaborate with Chris Haspell (Chris Haspell), who provided valuable guidance to both us and the client throughout the project.

We can’t reveal too many details yet, as this project remains confidential, but we can share that it’s set on a small island, is among the most picturesque sites we’ve worked on, and will feature a golf course designed by Clyde Johnson of Clyde Johnson. To bring the vision to life for potential investors, we created a cinematic teaser highlighting three holes that best capture the project's character. We were also delighted to collaborate with Chris Haspell (Chris Haspell), who provided valuable guidance to both us and the client throughout the project.

Harris Kalinka

12,046 次观看 • 6 个月前

Huge! mikecontango | τ,τ announced Project Rubicon at SmartCon. In partnership with Chainlink and building on Base. This is a DeFi primitive designed to unlock yield and bring serious liquidity to Bittensor. Here's what's happening: Project Rubicon is a liquid staked alpha bridge built by General Tensor in partnership with Chainlink CCIP. It's coming to Base network. What makes this unique? It's the first infrastructure connecting Bittensor subnet tokens to mainland EVM. Until now, Bittensor has been isolated, brilliant innovation stuck behind tools and complex wallet setups. Rubicon removes that barrier completely and enables access to the biggest innovation . Now retail users can find and trade alpha tokens directly on Aerodrome and through the Coinbase 🛡️ app. No more technical maze. Just a few clicks to join the decentralized AI revolution. The top subnets are already signed up as inaugural listing partners, seeding alpha liquidity from day one. This expands the Bittensor universe to where the real liquidity lives. And this is huge for #Bittensor. More utility More liquidity Easy accessibility

Huge! mikecontango | τ,τ announced Project Rubicon at SmartCon. In partnership with Chainlink and building on Base. This is a DeFi primitive designed to unlock yield and bring serious liquidity to Bittensor. Here's what's happening: Project Rubicon is a liquid staked alpha bridge built by General Tensor in partnership with Chainlink CCIP. It's coming to Base network. What makes this unique? It's the first infrastructure connecting Bittensor subnet tokens to mainland EVM. Until now, Bittensor has been isolated, brilliant innovation stuck behind tools and complex wallet setups. Rubicon removes that barrier completely and enables access to the biggest innovation . Now retail users can find and trade alpha tokens directly on Aerodrome and through the Coinbase 🛡️ app. No more technical maze. Just a few clicks to join the decentralized AI revolution. The top subnets are already signed up as inaugural listing partners, seeding alpha liquidity from day one. This expands the Bittensor universe to where the real liquidity lives. And this is huge for #Bittensor. More utility More liquidity Easy accessibility

Robin τ

10,810 次观看 • 8 个月前

✨ Personal AI is the next computing platform. AI is shifting from something you access to something you build with, locally, at the edge, and across systems. We’re unlocking new possibilities for developers: • AMD Ryzen AI Halo, a local-first developer system, preorder starting in June, develop AI without limits on your desk • Gorgon Halo with up to 192GB unified memory, supporting 300B+ parameter models locally, run massive models locally We’re excited to partner with clem 🤗🤗, Co-founder and CEO of Hugging Face, to advance open-source AI for Ryzen AI. Our focus is seamless AI, from model to deployment. Cloud, edge, device. One continuum. Multi-agent systems, local inference at scale, open models as infrastructure. This is the next computing era 🚀

✨ Personal AI is the next computing platform. AI is shifting from something you access to something you build with, locally, at the edge, and across systems. We’re unlocking new possibilities for developers: • AMD Ryzen AI Halo, a local-first developer system, preorder starting in June, develop AI without limits on your desk • Gorgon Halo with up to 192GB unified memory, supporting 300B+ parameter models locally, run massive models locally We’re excited to partner with clem 🤗🤗, Co-founder and CEO of Hugging Face, to advance open-source AI for Ryzen AI. Our focus is seamless AI, from model to deployment. Cloud, edge, device. One continuum. Multi-agent systems, local inference at scale, open models as infrastructure. This is the next computing era 🚀

Jack Huynh

47,822 次观看 • 2 个月前

NEWS: SpaceX has detailed Starmind, its plan for up to 1 million AI compute satellites in orbit, on a new website. Starmind moves AI data centers into space. Each satellite runs AI compute in orbit and beams results to Earth through Starlink lasers. The satellites draw constant solar power and cool for free in the vacuum. SpaceX filed that plan with the FCC in January. The first satellite, AI1, carries 150 kW of peak compute and spans 70 meters. Starship will carry 30 to 50 AI1 satellites per launch. SpaceX will build them at a new Gigasat factory in Bastrop, Texas. The AI chips come from a planned SpaceX and Tesla fab called Terafab. Two prototypes target early 2027, with thousands of satellites from late 2027.

NEWS: SpaceX has detailed Starmind, its plan for up to 1 million AI compute satellites in orbit, on a new website. Starmind moves AI data centers into space. Each satellite runs AI compute in orbit and beams results to Earth through Starlink lasers. The satellites draw constant solar power and cool for free in the vacuum. SpaceX filed that plan with the FCC in January. The first satellite, AI1, carries 150 kW of peak compute and spans 70 meters. Starship will carry 30 to 50 AI1 satellites per launch. SpaceX will build them at a new Gigasat factory in Bastrop, Texas. The AI chips come from a planned SpaceX and Tesla fab called Terafab. Two prototypes target early 2027, with thousands of satellites from late 2027.

Muskonomy

26,421 次观看 • 23 天前

How to Make AI Models More Useful? Discover the potential of AI and blockchain, and how features like RAG and TEEs can help NEAR unlock its full potential to build the world's best truly open AI. TL;DR - RAG (Retrieval-Augmented Generation) enhances AI agents by personalizing responses with relevant context. - Blockchain ensures transactions are processed transparently, efficiently, and securely. - NEAR Protocol is designed for AI, featuring a sharding architecture that supports high transaction volumes at low fees without compromising performance. It also offers privacy, user-owned data, and monetization opportunities. Study the NEAR stack, explore how to make AI models useful:

How to Make AI Models More Useful? Discover the potential of AI and blockchain, and how features like RAG and TEEs can help NEAR unlock its full potential to build the world's best truly open AI. TL;DR - RAG (Retrieval-Augmented Generation) enhances AI agents by personalizing responses with relevant context. - Blockchain ensures transactions are processed transparently, efficiently, and securely. - NEAR Protocol is designed for AI, featuring a sharding architecture that supports high transaction volumes at low fees without compromising performance. It also offers privacy, user-owned data, and monetization opportunities. Study the NEAR stack, explore how to make AI models useful:

NEARWEEK

23,509 次观看 • 1 年前