Loading video...

Video Failed to Load

Go Home

Centralized AI inference runs on one provider. NEAR AI's confidential inference draws from multiple compute partners distributed across regions instead of a single data center. No KYC required. No visibility into what you're running. Intelligence owned by the user.

NEARProtocol's profile picture

NEAR Protocol

1,828,823 subscribers

73,204 views • 1 month ago •via X (Twitter)

News & PoliticsHealth & WellnessScience & Technology
LIVE

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

George Zeng explains NEAR’s approach to user-owned AI. Confidential inference keeps your data yours. IronClaw runs agents where secrets never reach the model. NEAR Intents settles across chains. Intelligence, execution, settlement. The full stack, and it answers to you.
0:52

George Zeng explains NEAR’s approach to user-owned AI. Confidential inference keeps your data yours. IronClaw runs agents where secrets never reach the model. NEAR Intents settles across chains. Intelligence, execution, settlement. The full stack, and it answers to you.

NEAR Protocol

149,196 views • 19 days ago

AI inference is already live on UOMI Router. Access frontier open-source models through a decentralized inference network with: • Low-cost inference • OpenAI-compatible APIs • Verifiable compute • On-chain settlement Start building: And if you have idle GPUs, there's now a second way to participate. The UOMI Provider whitelist is officially open. Instead of letting your hardware sit unused, you can contribute GPU power to the UOMI Inference Network and earn from real AI demand. Join the Provider whitelist:
0:34

AI inference is already live on UOMI Router. Access frontier open-source models through a decentralized inference network with: • Low-cost inference • OpenAI-compatible APIs • Verifiable compute • On-chain settlement Start building: And if you have idle GPUs, there's now a second way to participate. The UOMI Provider whitelist is officially open. Instead of letting your hardware sit unused, you can contribute GPU power to the UOMI Inference Network and earn from real AI demand. Join the Provider whitelist:

Uomi

18,933 views • 1 month ago

OptimAI Core Node - More Than Mining. It’s Thinking. What does the OptimAI Core Node really do? + Crawls and indexes real-time web data + Runs lightweight AI inference tasks + Stores and retrieves data collaboratively + Powers decentralized search and compute Each node you run becomes part of a living AI infrastructure - one that learns, evolves, and grows smarter with every participant. The future of intelligence isn’t centralized. It’s distributed, and it starts with us. 👉 Join the network: 🎥 Watch the installation guide from our CCO, Ricardo, now for Windows users.
1:52

OptimAI Core Node - More Than Mining. It’s Thinking. What does the OptimAI Core Node really do? + Crawls and indexes real-time web data + Runs lightweight AI inference tasks + Stores and retrieves data collaboratively + Powers decentralized search and compute Each node you run becomes part of a living AI infrastructure - one that learns, evolves, and grows smarter with every participant. The future of intelligence isn’t centralized. It’s distributed, and it starts with us. 👉 Join the network: 🎥 Watch the installation guide from our CCO, Ricardo, now for Windows users.

OptimAI Network

243,232 views • 9 months ago

Lightchain AI – Decentralized Inference Demo 🍂🥧 + Blockchain Rewards In this video, we showcase how Lightchain AI’s decentralized inference engine (AIVM + PoI + Federated Learning) runs a real-world prompt “What is the best way to make pumpkin pie for thanksgiving dinner?” 💰 Blockchain Rewards in Action This demo doesn’t just show inference — it shows how inference providers are rewarded for their work. The provider that processed this prompt was automatically rewarded on the Lightchain testnet. 🍁 Why It Matters Unlike centralized AI, Lightchain AI distributes inference across a global network of participants. Nodes earn blockchain rewards for providing compute, ensuring a trustless, transparent, and economically sustainable ecosystem. With pay-to-prompt mechanics and DAO governance, Lightchain AI creates a fair AI economy for all. 🌐 Learn more at-
1:37

Lightchain AI – Decentralized Inference Demo 🍂🥧 + Blockchain Rewards In this video, we showcase how Lightchain AI’s decentralized inference engine (AIVM + PoI + Federated Learning) runs a real-world prompt “What is the best way to make pumpkin pie for thanksgiving dinner?” 💰 Blockchain Rewards in Action This demo doesn’t just show inference — it shows how inference providers are rewarded for their work. The provider that processed this prompt was automatically rewarded on the Lightchain testnet. 🍁 Why It Matters Unlike centralized AI, Lightchain AI distributes inference across a global network of participants. Nodes earn blockchain rewards for providing compute, ensuring a trustless, transparent, and economically sustainable ecosystem. With pay-to-prompt mechanics and DAO governance, Lightchain AI creates a fair AI economy for all. 🌐 Learn more at-

Lightchain AI

18,665 views • 11 months ago

Destra Edge — $DSYNC From Idea to Reality Yesterday, we introduced Destra Edge. Today, we show you what it means. For too long, AI has lived in data centers. Expensive. Centralized. Controlled by a few. Meanwhile, billions of smartphones sit idle. Powerful GPUs. Doing nothing. That doesn’t make sense. So we’re building something different. Destra Edge turns everyday smartphones into a global AI inference network. No servers. No cloud monopolies. Just real compute at the edge. This video shows the system taking shape — how inference is distributed, how it runs on-device, and how results come together without central control. This is not finished. And that’s the point. You’re watching it being built. Layer by layer. In public. The magic is in the scale. More phones means more power. More power means stronger AI. The network grows simply because people join. No passive mining. No fake activity. Only real GPU inference. Measured. Verified. Rewarded. AI shouldn’t live in warehouses. It should live in your pocket. This is edge-level decentralization. This is permissionless inference. This is Destra Network.
2:02

Destra Edge — $DSYNC From Idea to Reality Yesterday, we introduced Destra Edge. Today, we show you what it means. For too long, AI has lived in data centers. Expensive. Centralized. Controlled by a few. Meanwhile, billions of smartphones sit idle. Powerful GPUs. Doing nothing. That doesn’t make sense. So we’re building something different. Destra Edge turns everyday smartphones into a global AI inference network. No servers. No cloud monopolies. Just real compute at the edge. This video shows the system taking shape — how inference is distributed, how it runs on-device, and how results come together without central control. This is not finished. And that’s the point. You’re watching it being built. Layer by layer. In public. The magic is in the scale. More phones means more power. More power means stronger AI. The network grows simply because people join. No passive mining. No fake activity. Only real GPU inference. Measured. Verified. Rewarded. AI shouldn’t live in warehouses. It should live in your pocket. This is edge-level decentralization. This is permissionless inference. This is Destra Network.

Destra Network

20,156 views • 7 months ago

A glimpse into one of our core features 👀 Chat, Search, Compare, & Run #Inference for all AI models on ONE Platform: No more complicated setups, just seamless #AI magic. 💫 Be the first to experience - join waitlist now: #Inferium #AI #Inference
0:43

A glimpse into one of our core features 👀 Chat, Search, Compare, & Run #Inference for all AI models on ONE Platform: No more complicated setups, just seamless #AI magic. 💫 Be the first to experience - join waitlist now: #Inferium #AI #Inference

Inferium

1,079,270 views • 2 years ago

🎤 Pavel Durov, founder and CEO of Telegram Messenger, spoke at Blockchain Life 2026 2025 about Telegram Gifts, MiniApps success, and the dangers of centralized AI. Pavel announced Cocoon (Confidential Compute Open Network), a privacy-focused decentralized AI network powered by TON, launching in November. What it offers: 🔹 GPU owners can mine $TON by contributing computing power 🔹 App developers get secure inference without centralized data centers 🔹 Users experience AI services with complete privacy Apply: Watch the full speech 👇
15:43

🎤 Pavel Durov, founder and CEO of Telegram Messenger, spoke at Blockchain Life 2026 2025 about Telegram Gifts, MiniApps success, and the dangers of centralized AI. Pavel announced Cocoon (Confidential Compute Open Network), a privacy-focused decentralized AI network powered by TON, launching in November. What it offers: 🔹 GPU owners can mine $TON by contributing computing power 🔹 App developers get secure inference without centralized data centers 🔹 Users experience AI services with complete privacy Apply: Watch the full speech 👇

TON 💎

53,728 views • 9 months ago

One account. No borders. Your financial life, on-chain. Trade, access yield, and go confidential across 35+ networks from a single account, powered by NEAR. Get started:
0:28

One account. No borders. Your financial life, on-chain. Trade, access yield, and go confidential across 35+ networks from a single account, powered by NEAR. Get started:

NEARWEEK

16,449 views • 4 months ago

Distributed compute needs to be deployed with speed, scale, and sovereignty. Dan Wright joined Bloomberg Technology this week to talk about Galleon Forge One, our Series B, and the infrastructure buildout required to meet growing demand for inference, fine-tuning, and distributed model deployment at scale. The next wave of compute will not live only in centralized clouds. It will need to be deployed where land, energy, and data already exist. That is what Armada was built for. That is what Galleon delivers.
3:58

Distributed compute needs to be deployed with speed, scale, and sovereignty. Dan Wright joined Bloomberg Technology this week to talk about Galleon Forge One, our Series B, and the infrastructure buildout required to meet growing demand for inference, fine-tuning, and distributed model deployment at scale. The next wave of compute will not live only in centralized clouds. It will need to be deployed where land, energy, and data already exist. That is what Armada was built for. That is what Galleon delivers.

Armada

934,749 views • 2 months ago

Hyra AI Reborn. Stronger. Faster. Decentralized. 🔥 Reimagined. A new direction. A new era of decentralized intelligence. From privacy-first edge compute → to a global AI mesh owned by the people. From inference on your device → to digital sovereignty at scale. ✨ Every device is now a node. ✨ Every user is a Hyrasapien. ✨ Every action builds the future of AI. We’re not just rebranding ~ we’re reshaping the narrative of AI. Together, let’s build an AI infrastructure that’s secure, verifiable, and community-owned. 🔑 Your Device. Your Power. Your Sovereignty. 🌐 Explore the new era:
0:35

Hyra AI Reborn. Stronger. Faster. Decentralized. 🔥 Reimagined. A new direction. A new era of decentralized intelligence. From privacy-first edge compute → to a global AI mesh owned by the people. From inference on your device → to digital sovereignty at scale. ✨ Every device is now a node. ✨ Every user is a Hyrasapien. ✨ Every action builds the future of AI. We’re not just rebranding ~ we’re reshaping the narrative of AI. Together, let’s build an AI infrastructure that’s secure, verifiable, and community-owned. 🔑 Your Device. Your Power. Your Sovereignty. 🌐 Explore the new era:

Hyra AI

96,331 views • 10 months ago

INTERVIEW: AI, Surveillance, & the Fight for Digital Sovereignty | Near's Illia Polosukhin What happens when powerful AI requires an ID, governments can restrict access by country, and your most private data flows through centralized systems? NEAR Protocol co-founder Illia (root.near) (🇺🇦, ⋈) joins David Hoffman to explain why AI export controls could fracture the internet, how nationalized AI could become a surveillance layer, and why decentralized, user-owned AI may be the only real alternative. They also unpack Ironclaw, confidential inference, private agents, and NEAR’s push to make AI and assets fully sovereign. [TIMESTAMPS] 0:00 Intro 3:47 Pandora’s Box Opens 6:03 AI Meets Government 7:16 Decentralizing the Stack 8:51 Why Ban Fable? 11:18 Internet Balkanization 13:57 KYC for AI? 18:43 Nationalized Models 21:07 Building Private AI 26:30 Agent Marketplace 30:26 User-Owned Alignment 35:28 Winning Against Giants 40:19 Regulation Fuels Crypto 42:24 Global AI Adoption 45:05 Compute Is the Bottleneck 49:00 Enterprise Use Cases 51:39 Confidential TVL Explained 55:25 Full Sovereign Mode
57:26

INTERVIEW: AI, Surveillance, & the Fight for Digital Sovereignty | Near's Illia Polosukhin What happens when powerful AI requires an ID, governments can restrict access by country, and your most private data flows through centralized systems? NEAR Protocol co-founder Illia (root.near) (🇺🇦, ⋈) joins David Hoffman to explain why AI export controls could fracture the internet, how nationalized AI could become a surveillance layer, and why decentralized, user-owned AI may be the only real alternative. They also unpack Ironclaw, confidential inference, private agents, and NEAR’s push to make AI and assets fully sovereign. [TIMESTAMPS] 0:00 Intro 3:47 Pandora’s Box Opens 6:03 AI Meets Government 7:16 Decentralizing the Stack 8:51 Why Ban Fable? 11:18 Internet Balkanization 13:57 KYC for AI? 18:43 Nationalized Models 21:07 Building Private AI 26:30 Agent Marketplace 30:26 User-Owned Alignment 35:28 Winning Against Giants 40:19 Regulation Fuels Crypto 42:24 Global AI Adoption 45:05 Compute Is the Bottleneck 49:00 Enterprise Use Cases 51:39 Confidential TVL Explained 55:25 Full Sovereign Mode

Bankless

59,588 views • 1 month ago

Perplexity CEO Aravind Srinivas on the biggest threat to the data center industry: It's not competition. It's not regulation. It's decentralisation. "The biggest threat to a data center is if the intelligence can be packed locally on a chip that's running on the device and then there's no need to inference all of it on like one centralized data center." He outlines how this could work in practice. Personalisation doesn't necessarily require on-device model training. Retrieval augmented generation, tool calls, and local data can already tailor AI to individual users. But the real unlock? Test time training. Aravind Srinivas describes a future where AI lives on your device, watches how you work and gradually automates your repetitive tasks. "Imagine we crack test time training where the AI watches tasks you repeatedly do on your local system, adapts to you over time and starts automating a lot of the things you do." The key insight: in this model, the intelligence belongs to you. It's your data, your device, your personalised AI brain. And if that future arrives, the economics of centralised infrastructure start to collapse. "That really disrupts the whole data center industry. It doesn't make sense to spend all this money, 500 billion, 5 trillion, whatever on building all the centralized data centers across the world that do a lot of the intelligence workloads for people." The companies spending trillions on centralised infrastructure may want to rethink where intelligence actually needs to live.
1:15

Perplexity CEO Aravind Srinivas on the biggest threat to the data center industry: It's not competition. It's not regulation. It's decentralisation. "The biggest threat to a data center is if the intelligence can be packed locally on a chip that's running on the device and then there's no need to inference all of it on like one centralized data center." He outlines how this could work in practice. Personalisation doesn't necessarily require on-device model training. Retrieval augmented generation, tool calls, and local data can already tailor AI to individual users. But the real unlock? Test time training. Aravind Srinivas describes a future where AI lives on your device, watches how you work and gradually automates your repetitive tasks. "Imagine we crack test time training where the AI watches tasks you repeatedly do on your local system, adapts to you over time and starts automating a lot of the things you do." The key insight: in this model, the intelligence belongs to you. It's your data, your device, your personalised AI brain. And if that future arrives, the economics of centralised infrastructure start to collapse. "That really disrupts the whole data center industry. It doesn't make sense to spend all this money, 500 billion, 5 trillion, whatever on building all the centralized data centers across the world that do a lot of the intelligence workloads for people." The companies spending trillions on centralised infrastructure may want to rethink where intelligence actually needs to live.

Big Brain AI

90,241 views • 5 months ago

Weekend. The city slows down. OptimAI Network doesn’t. A query enters OptimAI Search.🔍 It doesn’t hit a single server. It fans out across decentralized nodes. Fetch. Cross-validate. Structure. Proof anchored. No black box ranking farm. No hidden data silos. Just distributed compute turning raw web data into verifiable intelligence for agents. Weekend for humans. Compounding for infrastructure. The future of search is not louder. It’s decentralized. 👉
1:59

Weekend. The city slows down. OptimAI Network doesn’t. A query enters OptimAI Search.🔍 It doesn’t hit a single server. It fans out across decentralized nodes. Fetch. Cross-validate. Structure. Proof anchored. No black box ranking farm. No hidden data silos. Just distributed compute turning raw web data into verifiable intelligence for agents. Weekend for humans. Compounding for infrastructure. The future of search is not louder. It’s decentralized. 👉

OptimAI Network

29,893 views • 6 months ago

NEW: Premium Inference 101 The Economics & Infrastructure Behind Running Trillion Parameter Models Rodrigo Liang, CEO & Co-Founder of SambaNova "Inference has arrived. 70-80% of those racks are running inference." "[Inference services] are generating lots of revenue, but not enough margin. In order for them to sustain, they've gotta be more profitable." "With SambaNova, that min quantum is down to 1 rack. Where if you have other service providers, [with] say, a DeepSeek model, now 1.5 trillion parameters, to run that, the min for some of the other providers might be 10-20 racks." SambaNova builds full-stack inference infrastructure. 16 chips to a 10kW air-cooled rack that runs trillion parameter models, where a GPU rack pulls 130kW. They just demonstrated the fastest MiniMax M2.7 inference in the world, as benchmarked by Artificial Analysis. The demo paired one NVIDIA H200 rack for prefill with one SambaRack SN50 for decode. Disaggregated inference: GPUs load the context, RDUs generate the tokens. Now serving JPMorgan, SoftBank, Saudi Aramco & DOE national labs, just valued at $11B on a $1B Series F led by General Atlantic. We Cover: › Why inference will need orders of magnitude more chips than training ever did › The 10kW rack vs the 130kW rack, & why air cooling decides geography › Running a 1T parameter model in one rack at full precision, no quantization › The agent latency problem: 20 agents, 2 seconds each, 40 seconds gone › Revenue per rack, & why inference providers have revenue but no margin › JPMorgan, sovereignty, & the move back to on-prem Filmed at the RAISE Summit in Paris. Thank you to Brex, MongoDB & AssemblyAI for helping make this trip & content series happen. 𝐓𝐈𝐌𝐄𝐒𝐓𝐀𝐌𝐏𝐒 (00:00) Rodrigo Liang , Co-Founder & CEO at SambaNova Systems (00:59) SambaNova’s Series F: $1B raise at an $11 billion valuation (03:00) The Inference problem nobody saw coming (04:52) SambaNova's chip evolution (07:19) Running a trillion-parameter model on a single rack (11:00) Do $100 billion data centers actually make sense? (14:14) What "premium inference" really means (18:28) Speed is about to become AI's biggest price tag (20:43) Starlink, edge computing, & AI reaching every corner of the planet (24:27) Working alongside NVIDIA & rival chipmakers (27:49) How customers actually measure inference performance (32:07) The biggest bottlenecks in AI's global land grab (35:12) Justifying the billion-dollar AI valuations (37:53) Why SambaNova refuses to build its own cloud (41:03) The "AI sovereignty" debate (43:48) Data privacy fears are driving the return to on-prem AI (47:55) How to actually get ROI out of AI spend (51:16) The one question every business should be asking about AI (56:09) The mentors & lessons behind a 32-year career in chips (58:02) Unveiling SambaNova's newest chip, the SN50
59:56

NEW: Premium Inference 101 The Economics & Infrastructure Behind Running Trillion Parameter Models Rodrigo Liang, CEO & Co-Founder of SambaNova "Inference has arrived. 70-80% of those racks are running inference." "[Inference services] are generating lots of revenue, but not enough margin. In order for them to sustain, they've gotta be more profitable." "With SambaNova, that min quantum is down to 1 rack. Where if you have other service providers, [with] say, a DeepSeek model, now 1.5 trillion parameters, to run that, the min for some of the other providers might be 10-20 racks." SambaNova builds full-stack inference infrastructure. 16 chips to a 10kW air-cooled rack that runs trillion parameter models, where a GPU rack pulls 130kW. They just demonstrated the fastest MiniMax M2.7 inference in the world, as benchmarked by Artificial Analysis. The demo paired one NVIDIA H200 rack for prefill with one SambaRack SN50 for decode. Disaggregated inference: GPUs load the context, RDUs generate the tokens. Now serving JPMorgan, SoftBank, Saudi Aramco & DOE national labs, just valued at $11B on a $1B Series F led by General Atlantic. We Cover: › Why inference will need orders of magnitude more chips than training ever did › The 10kW rack vs the 130kW rack, & why air cooling decides geography › Running a 1T parameter model in one rack at full precision, no quantization › The agent latency problem: 20 agents, 2 seconds each, 40 seconds gone › Revenue per rack, & why inference providers have revenue but no margin › JPMorgan, sovereignty, & the move back to on-prem Filmed at the RAISE Summit in Paris. Thank you to Brex, MongoDB & AssemblyAI for helping make this trip & content series happen. 𝐓𝐈𝐌𝐄𝐒𝐓𝐀𝐌𝐏𝐒 (00:00) Rodrigo Liang , Co-Founder & CEO at SambaNova Systems (00:59) SambaNova’s Series F: $1B raise at an $11 billion valuation (03:00) The Inference problem nobody saw coming (04:52) SambaNova's chip evolution (07:19) Running a trillion-parameter model on a single rack (11:00) Do $100 billion data centers actually make sense? (14:14) What "premium inference" really means (18:28) Speed is about to become AI's biggest price tag (20:43) Starlink, edge computing, & AI reaching every corner of the planet (24:27) Working alongside NVIDIA & rival chipmakers (27:49) How customers actually measure inference performance (32:07) The biggest bottlenecks in AI's global land grab (35:12) Justifying the billion-dollar AI valuations (37:53) Why SambaNova refuses to build its own cloud (41:03) The "AI sovereignty" debate (43:48) Data privacy fears are driving the return to on-prem AI (47:55) How to actually get ROI out of AI spend (51:16) The one question every business should be asking about AI (56:09) The mentors & lessons behind a 32-year career in chips (58:02) Unveiling SambaNova's newest chip, the SN50

Molly O’Shea

134,960 views • 1 month ago

AI is rapidly evolving, and it’s reshaping how and where computing happens. From data center infrastructure to AI systems in the physical world, to the future of personal AI devices, intelligence is being distributed across a new compute landscape. At the edge, more agentic experiences are starting to show up, expanding what devices need to do. Tomorrow in New York, we’ll share what Qualcomm is building across it and how it will drive our next phase of growth. Stay tuned for #QCOMInvestorDay.
0:32

AI is rapidly evolving, and it’s reshaping how and where computing happens. From data center infrastructure to AI systems in the physical world, to the future of personal AI devices, intelligence is being distributed across a new compute landscape. At the edge, more agentic experiences are starting to show up, expanding what devices need to do. Tomorrow in New York, we’ll share what Qualcomm is building across it and how it will drive our next phase of growth. Stay tuned for #QCOMInvestorDay.

Cristiano R. Amon

15,783 views • 1 month ago

Think of OptimAI as a living network - hundred thousands of nodes sensing the web, computing knowledge, verifying truth. Lite Nodes → surface signals & activity Edge Nodes → compute at the edge, background mining Core Nodes → crawl, embed, index, inference, validate & serve workloads Distributed across 179 countries, approaching 800,000 total installations and more than 6,000 Core Nodes online. Every node strengthens collective intelligence. Every contribution expands the data layer powering Agentic AI. This is AI not controlled, but co-created. Built for the People, by the People. Start your node and take part in the next era of intelligence:
0:20

Think of OptimAI as a living network - hundred thousands of nodes sensing the web, computing knowledge, verifying truth. Lite Nodes → surface signals & activity Edge Nodes → compute at the edge, background mining Core Nodes → crawl, embed, index, inference, validate & serve workloads Distributed across 179 countries, approaching 800,000 total installations and more than 6,000 Core Nodes online. Every node strengthens collective intelligence. Every contribution expands the data layer powering Agentic AI. This is AI not controlled, but co-created. Built for the People, by the People. Start your node and take part in the next era of intelligence:

OptimAI Network

19,365 views • 8 months ago

70,000 Phones, One AI Agent — The World's Largest Edge AI Fleet Runs on Hermes We turned 70,000 phones into a shared AI compute network. Any device owner contributes idle compute. Any developer taps distributed inference at a fraction of cloud cost. Not a concept. Not a whitepaper. 70K devices online today. The problem: orchestrating a shared network of heterogeneous edge devices — different chipsets, different memory, different thermal profiles, different owners — is a coordination nightmare no human team can handle manually. So we gave the network a brain: Nous Research Hermes Agent. Hermes connects to 16 MCP servers and runs 24/7: 🔬 Research Loop — Tracks every breakthrough in on-device inference: quantization (GPTQ/AWQ/GGUF), speculative decoding on mobile SoCs, federated learning protocols. Auto-imports papers into NotebookLM. 36 research topics, zero manual curation. 🌐 Network Intelligence — Monitors device availability, compute capacity, and workload distribution across the shared fleet. Surfaces bottlenecks before they cascade. 🧬 Tech Tree Optimizer — Maps the full optimization frontier: from KV-cache compression to on-device LoRA to peer-to-peer model sharding. Hermes autonomously identifies which research paths unlock the most network-wide throughput gains. The result: a self-improving shared compute network. Research compounds daily. The fleet gets smarter without human intervention. Cloud AI scales with money. We scale with people. #HermesHackathon Teknium 🪽 Delphi Digital Tommy
0:30

70,000 Phones, One AI Agent — The World's Largest Edge AI Fleet Runs on Hermes We turned 70,000 phones into a shared AI compute network. Any device owner contributes idle compute. Any developer taps distributed inference at a fraction of cloud cost. Not a concept. Not a whitepaper. 70K devices online today. The problem: orchestrating a shared network of heterogeneous edge devices — different chipsets, different memory, different thermal profiles, different owners — is a coordination nightmare no human team can handle manually. So we gave the network a brain: Nous Research Hermes Agent. Hermes connects to 16 MCP servers and runs 24/7: 🔬 Research Loop — Tracks every breakthrough in on-device inference: quantization (GPTQ/AWQ/GGUF), speculative decoding on mobile SoCs, federated learning protocols. Auto-imports papers into NotebookLM. 36 research topics, zero manual curation. 🌐 Network Intelligence — Monitors device availability, compute capacity, and workload distribution across the shared fleet. Surfaces bottlenecks before they cascade. 🧬 Tech Tree Optimizer — Maps the full optimization frontier: from KV-cache compression to on-device LoRA to peer-to-peer model sharding. Hermes autonomously identifies which research paths unlock the most network-wide throughput gains. The result: a self-improving shared compute network. Research compounds daily. The fleet gets smarter without human intervention. Cloud AI scales with money. We scale with people. #HermesHackathon Teknium 🪽 Delphi Digital Tommy

Oyster Republic 🦪📲🦞👓

20,721 views • 5 months ago

🚨 I DON'T KNOW WHAT PEOPLE ARE WAITING FOR! This guy just deployed a full GPU compute farm in a data center Not renting cloud instances, not paying per-token, just racks of consumer GPUs on warehouse shelving doing AI inference 24/7 The setup is wild: open-frame rigs with rows of white fans, PCIe riser cables, server-grade PSUs, and a PDU for power distribution. All mounted on red industrial shelving with raised-floor cooling underneath. It looks like a crypto mining op, because it probably was one Here's what's actually happening: > Ex-mining rigs are being redeployed for AI inference at scale > Each rig runs 6-8 GPUs on PCIe risers connected to a single motherboard > Fans pull cold air from the raised floor through the open frames > A PDU on each rack handles power distribution across all rigs > The whole thing runs in a colocation facility with cheap power rates The economics make sense: cloud GPU pricing runs $2-4/hr per card. Own the hardware, colocate it for $0.08-0.12/kWh, and you're printing compute at a fraction of the cost The crypto-to-AI pipeline is real. Same hardware, same facilities, same power contracts, completely different revenue model Bookmark this!
0:53

🚨 I DON'T KNOW WHAT PEOPLE ARE WAITING FOR! This guy just deployed a full GPU compute farm in a data center Not renting cloud instances, not paying per-token, just racks of consumer GPUs on warehouse shelving doing AI inference 24/7 The setup is wild: open-frame rigs with rows of white fans, PCIe riser cables, server-grade PSUs, and a PDU for power distribution. All mounted on red industrial shelving with raised-floor cooling underneath. It looks like a crypto mining op, because it probably was one Here's what's actually happening: > Ex-mining rigs are being redeployed for AI inference at scale > Each rig runs 6-8 GPUs on PCIe risers connected to a single motherboard > Fans pull cold air from the raised floor through the open frames > A PDU on each rack handles power distribution across all rigs > The whole thing runs in a colocation facility with cheap power rates The economics make sense: cloud GPU pricing runs $2-4/hr per card. Own the hardware, colocate it for $0.08-0.12/kWh, and you're printing compute at a fraction of the cost The crypto-to-AI pipeline is real. Same hardware, same facilities, same power contracts, completely different revenue model Bookmark this!

cristal

16,305 views • 1 month ago

AI is leaving the data center. NVIDIA is betting on local, personal compute. Edge-agent research is showing why that matters: the closer the agent is to the user, the more usable it becomes. That is the OptimAI Claw thesis. Search brings the signal. Claw turns it into action on your device. The network keeps intelligence distributed. This is what agent-native infrastructure looks like when it is built for the edge.
1:15

AI is leaving the data center. NVIDIA is betting on local, personal compute. Edge-agent research is showing why that matters: the closer the agent is to the user, the more usable it becomes. That is the OptimAI Claw thesis. Search brings the signal. Claw turns it into action on your device. The network keeps intelligence distributed. This is what agent-native infrastructure looks like when it is built for the edge.

OptimAI Network

27,462 views • 2 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

37,555 views • 4 months ago