Loading video...

Video Failed to Load

Go Home

Most speculative decoding still drafts tokens one at a time. That's not parallel generation — it just hides the serial loop behind a smaller model. UC San Diego's z-lab just drew a clear line between the two. They released DFlash — a lightweight block diffusion model that drafts a...

Marktechpost's profile picture

Marktechpost AI

11,515 subscribers

23,328 views • 1 month ago •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

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

A tricky LLM interview question: You're serving a reasoning model on vLLM, and it keeps running out of GPU memory on long traces. So you add KV cache compression and evict 90% of the cached tokens. VRAM usage stays as is and GPU still runs out of memory. Why? (answer below) Evicting 90% of the KV cache can free almost none of the memory it was using. This sounds counterintuitive, but it follows directly from how production servers store the cache today. The KV cache grows with every token a model generates. Each token appends its key and value vectors across every layer, and nothing is freed while generation continues. This is the dominant memory cost for reasoning models. If a 32K-token CoT caches ~32K tokens of KV vectors, a Qwen3-32B with 4-bit weights will run out-of-memory around 24K tokens on a 24GB GPU. One obvious solution is to keep the important tokens and drop the rest, since attention is sparse enough to allow it. But this does not solve the memory problem yet. The reason is paged attention, which is the memory manager behind vLLM and most production servers. Under the hood, it splits GPU memory into fixed physical blocks, each one holds the KV for about 16 tokens. This block returns to the allocator only when every slot inside it is empty. Since the eviction logic selects tokens by importance, and such tokens are scattered across blocks... ...so despite eviction, almost every block is left with at least some survivor tokens. For instance, if the logic evicts 14k of 16k tokens across 1,000 blocks, most likely every block will still have a token. This means the allocator frees almost nothing. Placing the new tokens into those freed slots is not ideal because it breaks the cache's layout. Say token 16,001 arrives, and it's placed in the slot the 40th token used to hold. The cache now reads position 38, then 16,001, then 41, so the cache is no longer in token order. Attention can still compute the right answer from that, but only if every slot now carries a separate note recording which position it actually holds. This introduces another bookkeeping cost that an in-order layout inherently avoids. So the cache is logically 90% smaller and still physically the same size. Many compression results miss this because they measure on pre-allocated contiguous tensors rather than a paged server. There's another problem. Eviction methods pick which tokens to keep by looking at the attention scores themselves (as expected). But fast attention kernels used in production, like FlashAttention, never save those scores. They compute attention in small pieces and throw the full score grid away as they go, which is also why they're fast. So the exact signal eviction methods need isn't available in memory. The workaround is to fall back to eager attention and build the full matrix, which gives up the speed FlashAttention was there to provide. NVIDIA published a method called TriAttention to solve both these problems. It never needs attention scores. Instead, it scores tokens from the geometry of the model's key and query vectors before RoPE is applied, where those vectors sit in stable clusters. For the memory problem, it runs a compaction pass every 128 decoded tokens. The surviving tokens slide forward to close the holes eviction creates, so whole blocks empty out and return to the allocator while the cache stays in token order. On long reasoning traces, the approach matches full-attention accuracy while decoding 2.5x faster and using 10.7x less KV memory. KV cache compression is a big infrastructure problem. The number that decides whether it works is the count of freed blocks, not the count of evicted tokens. You can find the NVIDIA write-up here: I wrote a first-principles breakdown of how the KV cache works. It walks through why the model stores keys and values at all, why the cache grows with every token, and a comparison of LLM generation speed with and without KV caching. Read it below.
0:23

A tricky LLM interview question: You're serving a reasoning model on vLLM, and it keeps running out of GPU memory on long traces. So you add KV cache compression and evict 90% of the cached tokens. VRAM usage stays as is and GPU still runs out of memory. Why? (answer below) Evicting 90% of the KV cache can free almost none of the memory it was using. This sounds counterintuitive, but it follows directly from how production servers store the cache today. The KV cache grows with every token a model generates. Each token appends its key and value vectors across every layer, and nothing is freed while generation continues. This is the dominant memory cost for reasoning models. If a 32K-token CoT caches ~32K tokens of KV vectors, a Qwen3-32B with 4-bit weights will run out-of-memory around 24K tokens on a 24GB GPU. One obvious solution is to keep the important tokens and drop the rest, since attention is sparse enough to allow it. But this does not solve the memory problem yet. The reason is paged attention, which is the memory manager behind vLLM and most production servers. Under the hood, it splits GPU memory into fixed physical blocks, each one holds the KV for about 16 tokens. This block returns to the allocator only when every slot inside it is empty. Since the eviction logic selects tokens by importance, and such tokens are scattered across blocks... ...so despite eviction, almost every block is left with at least some survivor tokens. For instance, if the logic evicts 14k of 16k tokens across 1,000 blocks, most likely every block will still have a token. This means the allocator frees almost nothing. Placing the new tokens into those freed slots is not ideal because it breaks the cache's layout. Say token 16,001 arrives, and it's placed in the slot the 40th token used to hold. The cache now reads position 38, then 16,001, then 41, so the cache is no longer in token order. Attention can still compute the right answer from that, but only if every slot now carries a separate note recording which position it actually holds. This introduces another bookkeeping cost that an in-order layout inherently avoids. So the cache is logically 90% smaller and still physically the same size. Many compression results miss this because they measure on pre-allocated contiguous tensors rather than a paged server. There's another problem. Eviction methods pick which tokens to keep by looking at the attention scores themselves (as expected). But fast attention kernels used in production, like FlashAttention, never save those scores. They compute attention in small pieces and throw the full score grid away as they go, which is also why they're fast. So the exact signal eviction methods need isn't available in memory. The workaround is to fall back to eager attention and build the full matrix, which gives up the speed FlashAttention was there to provide. NVIDIA published a method called TriAttention to solve both these problems. It never needs attention scores. Instead, it scores tokens from the geometry of the model's key and query vectors before RoPE is applied, where those vectors sit in stable clusters. For the memory problem, it runs a compaction pass every 128 decoded tokens. The surviving tokens slide forward to close the holes eviction creates, so whole blocks empty out and return to the allocator while the cache stays in token order. On long reasoning traces, the approach matches full-attention accuracy while decoding 2.5x faster and using 10.7x less KV memory. KV cache compression is a big infrastructure problem. The number that decides whether it works is the count of freed blocks, not the count of evicted tokens. You can find the NVIDIA write-up here: I wrote a first-principles breakdown of how the KV cache works. It walks through why the model stores keys and values at all, why the cache grows with every token, and a comparison of LLM generation speed with and without KV caching. Read it below.

Avi Chawla

268,010 views • 27 days ago

NVIDIA just removed one of the biggest friction points in Voice AI. PersonaPlex-7B is an open-source, full-duplex conversational model. Free, open source (MIT), with open model weights on Hugging Face 🤗 Links to repo and weights in 🧵↓ The traditional ASR → LLM → TTS pipeline forces rigid turn-taking. It’s efficient, but it never feels natural. PersonaPlex-7B changes that. This NVIDIA model can listen and speak at the same time. It runs directly on continuous audio tokens with a dual-stream transformer, generating text and audio in parallel instead of passing control between components. That unlocks: → instant back-channel responses → interruptions that feel human → real conversational rhythm Persona control is fully zero-shot! If you’re building low-latency assistants or support agents, this is a big step forward 🔥
0:42

NVIDIA just removed one of the biggest friction points in Voice AI. PersonaPlex-7B is an open-source, full-duplex conversational model. Free, open source (MIT), with open model weights on Hugging Face 🤗 Links to repo and weights in 🧵↓ The traditional ASR → LLM → TTS pipeline forces rigid turn-taking. It’s efficient, but it never feels natural. PersonaPlex-7B changes that. This NVIDIA model can listen and speak at the same time. It runs directly on continuous audio tokens with a dual-stream transformer, generating text and audio in parallel instead of passing control between components. That unlocks: → instant back-channel responses → interruptions that feel human → real conversational rhythm Persona control is fully zero-shot! If you’re building low-latency assistants or support agents, this is a big step forward 🔥

Charly Wargnier

565,090 views • 6 months ago

I had to test it myself to believe this unreal inference speed. 3,000 tokens/s for 1 user on standard datacenter GPUs. They leveraged a hidden efficiency gap in how GPUs generate tokens. Kog just achieved 3,000 tokens/s on 8× AMD MI300X GPUs and 2,100 on 8× NVIDIA H200 (FP16, no speculative decoding). Their tech preview is on a 2B model, and they show how their techniques will scale to large frontier MoE models at similar speeds. That's a huge number because normal low-batch GPU decoding for 2B to 8B models is usually closer to 100 to 300 tokens/s per request, so Kog is claiming something like a 10X to 30X jump in the speed one user actually feels. Their trick: they are getting the speed by treating LLM decoding as a memory streaming problem, not mainly a math problem. For 1 user at batch size 1, the GPU is not doing big, efficient matrix-matrix work like in training or large-batch serving; it is repeatedly pulling the model’s active weights from high-bandwidth memory for each new token, so speed depends on how smoothly those weights keep flowing. Normal inference stacks keep breaking that flow. They run many separate GPU programs for different parts of the model, move intermediate results through memory, wait at synchronization points, talk back to the CPU for scheduling or sampling, and then repeat this token after token. Kog’s answer is to co-design 3 things that are usually tuned separately: the runtime, the low-level GPU code, and the model architecture. The biggest engineering move is the monokernel, where the whole decode pass runs as 1 persistent GPU-resident program, including sampling, so the system does not keep stopping for kernel launches, CPU scheduling, and intermediate memory round trips. They also rebuilt synchronization, because their own measurements say grid sync was eating around 35% of token-generation time; instead of making every compute unit wait at a broad barrier, each unit waits only for the exact data it needs. On AMD MI300X, they also map memory access around the chiplet layout, because memory latency changes depending on which die makes the request. Then their Laneformer model uses Delayed Tensor Parallelism, which lets cross-GPU communication happen in the background instead of blocking every layer.
1:08

I had to test it myself to believe this unreal inference speed. 3,000 tokens/s for 1 user on standard datacenter GPUs. They leveraged a hidden efficiency gap in how GPUs generate tokens. Kog just achieved 3,000 tokens/s on 8× AMD MI300X GPUs and 2,100 on 8× NVIDIA H200 (FP16, no speculative decoding). Their tech preview is on a 2B model, and they show how their techniques will scale to large frontier MoE models at similar speeds. That's a huge number because normal low-batch GPU decoding for 2B to 8B models is usually closer to 100 to 300 tokens/s per request, so Kog is claiming something like a 10X to 30X jump in the speed one user actually feels. Their trick: they are getting the speed by treating LLM decoding as a memory streaming problem, not mainly a math problem. For 1 user at batch size 1, the GPU is not doing big, efficient matrix-matrix work like in training or large-batch serving; it is repeatedly pulling the model’s active weights from high-bandwidth memory for each new token, so speed depends on how smoothly those weights keep flowing. Normal inference stacks keep breaking that flow. They run many separate GPU programs for different parts of the model, move intermediate results through memory, wait at synchronization points, talk back to the CPU for scheduling or sampling, and then repeat this token after token. Kog’s answer is to co-design 3 things that are usually tuned separately: the runtime, the low-level GPU code, and the model architecture. The biggest engineering move is the monokernel, where the whole decode pass runs as 1 persistent GPU-resident program, including sampling, so the system does not keep stopping for kernel launches, CPU scheduling, and intermediate memory round trips. They also rebuilt synchronization, because their own measurements say grid sync was eating around 35% of token-generation time; instead of making every compute unit wait at a broad barrier, each unit waits only for the exact data it needs. On AMD MI300X, they also map memory access around the chiplet layout, because memory latency changes depending on which die makes the request. Then their Laneformer model uses Delayed Tensor Parallelism, which lets cross-GPU communication happen in the background instead of blocking every layer.

Rohan Paul

13,148 views • 1 month ago

SGLang now supports DSpark, enabling confidence-driven, variable-length verification for speculative decoding 🎉 DSpark addresses a key bottleneck under load: instead of verifying every draft token, it verifies only where the draft model is confident, so the gains hold even as batch size scales. We heavily optimized variable-length verification in SGLang. Across batch sizes 1 to 256, DSpark gives the best throughput/latency tradeoff on DeepSeek-V4-Flash, ahead of both MTP and non-spec. At high concurrency, dynamic scheduling provides up to ~20% higher throughput compared to a fixed budget, while maintaining high verification quality across workloads. With fused kernels and zero-overhead scheduling, DeepSeek-V4-Pro reaches 383.7 tok/s at B=1 on B300. DSpark is now available in SGLang with support for Qwen3 and DeepSeek-V4. Thanks DeepSeek for open-sourcing! Blog with full technical details and commands to run below 👇
0:33

SGLang now supports DSpark, enabling confidence-driven, variable-length verification for speculative decoding 🎉 DSpark addresses a key bottleneck under load: instead of verifying every draft token, it verifies only where the draft model is confident, so the gains hold even as batch size scales. We heavily optimized variable-length verification in SGLang. Across batch sizes 1 to 256, DSpark gives the best throughput/latency tradeoff on DeepSeek-V4-Flash, ahead of both MTP and non-spec. At high concurrency, dynamic scheduling provides up to ~20% higher throughput compared to a fixed budget, while maintaining high verification quality across workloads. With fused kernels and zero-overhead scheduling, DeepSeek-V4-Pro reaches 383.7 tok/s at B=1 on B300. DSpark is now available in SGLang with support for Qwen3 and DeepSeek-V4. Thanks DeepSeek for open-sourcing! Blog with full technical details and commands to run below 👇

LMSYS Org

167,269 views • 18 days ago

TWO BOXES THE SIZE OF A MAC MINI JUST RAN A 235 BILLION PARAMETER MODEL ON A DESK It is two NVIDIA DGX Spark units linked by a single cable. A year ago a model this size meant renting a GPU cluster by the hour. Now it sits next to your monitor for around $8,000. Here is the twist most people miss. Linking them does not create one shared 256GB memory pool. The model is split across both boxes, and that is the only reason a 235B model fits at all. It answers at roughly 10 tokens per second, and both chips sit at just 74 degrees while sipping around 50 watts. Every token stays on the desk. Nothing touches a cloud, and nothing leaves the room. The ceiling for what you can run at home just jumped from 70B to 235B. Bookmark this & Watch it run ↓
1:21

TWO BOXES THE SIZE OF A MAC MINI JUST RAN A 235 BILLION PARAMETER MODEL ON A DESK It is two NVIDIA DGX Spark units linked by a single cable. A year ago a model this size meant renting a GPU cluster by the hour. Now it sits next to your monitor for around $8,000. Here is the twist most people miss. Linking them does not create one shared 256GB memory pool. The model is split across both boxes, and that is the only reason a 235B model fits at all. It answers at roughly 10 tokens per second, and both chips sit at just 74 degrees while sipping around 50 watts. Every token stays on the desk. Nothing touches a cloud, and nothing leaves the room. The ceiling for what you can run at home just jumped from 70B to 235B. Bookmark this & Watch it run ↓

slash1s

101,133 views • 1 month ago

Nvidia is pulling off the most sophisticated financial loop in tech history. They invested $40 BILLION in its own customers in just 5 months. Here's why this could blow up the entire AI economy: Nvidia generated $97 billion in free cash flow last year. Instead of sitting on it, Jensen started writing checks to every company in the AI supply chain. Not small checks. We're talking about billions at a time. And almost every single one of those companies turns around and spends that money on Nvidia chips. Follow the money: $30 billion into OpenAI. OpenAI is one of Nvidia's largest GPU customers and spends billions annually on Nvidia hardware through cloud providers. $2 billion into CoreWeave, a company that exists exclusively to rent out data centers full of Nvidia GPUs. $2 billion into Marvell for silicon photonics that connects Nvidia systems. $2 billion into Lumentum for optical tech that powers Nvidia data centers. $2 billion into Coherent for the same thing. $2 billion into Nebius, an AI cloud company deploying Nvidia infrastructure. $3.2 billion into Corning, the glassmaker building three new US factories specifically to make fiber optic cables for Nvidia's next-gen systems. $2.1 billion into IREN, a data center operator that just agreed to deploy 5 gigawatts of Nvidia-designed infrastructure. And the list goes on. Every single recipient either buys Nvidia chips directly, builds infrastructure that runs on Nvidia chips, or manufactures components that go inside Nvidia systems. Matthew Bryson, an analyst at Wedbush Securities, said in a research note that Nvidia's dealmaking fits "squarely into the circular investment theme." Bloomberg even published an entire interactive feature this week titled "AI Circular Deals: How Microsoft, OpenAI and Nvidia Keep Paying Each Other." The piece maps how capital flows between the same handful of companies and gets counted as revenue multiple times along the way. But here's the part that makes this genuinely complicated: Nvidia's $5 billion investment in Intel from September is now worth over $25 billion. That's a 5x return in months. Their private company portfolio went from $3.4 billion to $22.3 billion on the balance sheet in a single year. They booked $8.9 billion in gains from equity investments alone. So when critics say "circular investing," Nvidia can point to Intel and say "we turned $5 billion into $25 billion, this is just smart capital deployment." And they're not wrong. Some of these bets ARE paying off like crazy. The real question is whether Nvidia is a chipmaker that happens to invest, or a venture fund that happens to sell chips. Because right now Jensen is doing both at a scale that has never existed in the semiconductor industry. No chipmaker in history has EVER invested $40 billion in its own ecosystem in five months. Last fiscal year Nvidia invested $17.5 billion in private companies. Their SEC filing literally says those investments include "AI model companies that purchase its products directly or through cloud service providers." They're saying it themselves: We invest in companies that buy our products. On Nvidia's last earnings call, Jensen told investors their investments are focused on "expanding and deepening our ecosystem reach." Translate that from CEO-speak and it means " we're funding the companies that fund us. The bull case says Nvidia is building an unbreakable moat by financing the entire AI supply chain and ensuring it all runs on Nvidia hardware. The bear case says this is the most elaborate circular revenue scheme since the subprime mortgage era and it all breaks apart the moment one domino falls. Both cases use the exact same evidence.
1:01

Nvidia is pulling off the most sophisticated financial loop in tech history. They invested $40 BILLION in its own customers in just 5 months. Here's why this could blow up the entire AI economy: Nvidia generated $97 billion in free cash flow last year. Instead of sitting on it, Jensen started writing checks to every company in the AI supply chain. Not small checks. We're talking about billions at a time. And almost every single one of those companies turns around and spends that money on Nvidia chips. Follow the money: $30 billion into OpenAI. OpenAI is one of Nvidia's largest GPU customers and spends billions annually on Nvidia hardware through cloud providers. $2 billion into CoreWeave, a company that exists exclusively to rent out data centers full of Nvidia GPUs. $2 billion into Marvell for silicon photonics that connects Nvidia systems. $2 billion into Lumentum for optical tech that powers Nvidia data centers. $2 billion into Coherent for the same thing. $2 billion into Nebius, an AI cloud company deploying Nvidia infrastructure. $3.2 billion into Corning, the glassmaker building three new US factories specifically to make fiber optic cables for Nvidia's next-gen systems. $2.1 billion into IREN, a data center operator that just agreed to deploy 5 gigawatts of Nvidia-designed infrastructure. And the list goes on. Every single recipient either buys Nvidia chips directly, builds infrastructure that runs on Nvidia chips, or manufactures components that go inside Nvidia systems. Matthew Bryson, an analyst at Wedbush Securities, said in a research note that Nvidia's dealmaking fits "squarely into the circular investment theme." Bloomberg even published an entire interactive feature this week titled "AI Circular Deals: How Microsoft, OpenAI and Nvidia Keep Paying Each Other." The piece maps how capital flows between the same handful of companies and gets counted as revenue multiple times along the way. But here's the part that makes this genuinely complicated: Nvidia's $5 billion investment in Intel from September is now worth over $25 billion. That's a 5x return in months. Their private company portfolio went from $3.4 billion to $22.3 billion on the balance sheet in a single year. They booked $8.9 billion in gains from equity investments alone. So when critics say "circular investing," Nvidia can point to Intel and say "we turned $5 billion into $25 billion, this is just smart capital deployment." And they're not wrong. Some of these bets ARE paying off like crazy. The real question is whether Nvidia is a chipmaker that happens to invest, or a venture fund that happens to sell chips. Because right now Jensen is doing both at a scale that has never existed in the semiconductor industry. No chipmaker in history has EVER invested $40 billion in its own ecosystem in five months. Last fiscal year Nvidia invested $17.5 billion in private companies. Their SEC filing literally says those investments include "AI model companies that purchase its products directly or through cloud service providers." They're saying it themselves: We invest in companies that buy our products. On Nvidia's last earnings call, Jensen told investors their investments are focused on "expanding and deepening our ecosystem reach." Translate that from CEO-speak and it means " we're funding the companies that fund us. The bull case says Nvidia is building an unbreakable moat by financing the entire AI supply chain and ensuring it all runs on Nvidia hardware. The bear case says this is the most elaborate circular revenue scheme since the subprime mortgage era and it all breaks apart the moment one domino falls. Both cases use the exact same evidence.

Ricardo

159,113 views • 2 months ago

Inside Nemotron and NVIDIA's AI lab: my conversation with Bryan Catanzaro (Bryan Catanzaro). NVIDIA is a chip company. So why does it put hundreds of researchers on building AI models - and then give them away for free? We go deep into the Nemotron models, what it takes to build a top AI lab, and the future of frontier AI. 01:33 - Is open source AI catching the frontier? 05:29 - Do closed labs blocking distillation slow open source down? 07:42 - Is the US falling behind China? 10:30 - Why companies actually choose open models 12:39 - A "crazy" 2008 bet: machine learning on GPUs 15:33 - Working with Andrew Ng and Dario Amodei at Baidu 17:41 - Coming back to NVIDIA: DLSS and the birth of Megatron 21:55 - The real reason NVIDIA builds its own models 24:28 - Is Moore's Law really dead? 33:37 - The Nemotron family: Nano, Super, Ultra 35:09 - Built for agents: why NVIDIA bets on speed 36:02 - How you train a 550B model in 4 bits 39:25 - Hybrid Mamba-Transformer, explained simply 42:31 - Mixture of experts, and why NVIDIA built NVL72 around it 47:26 - Why a 1-million-token context window matters 49:26 - Multi-token prediction: how the model predicts 5 tokens at once 52:47 - Multi-teacher distillation: teaching one model from many 58:01 - Where reinforcement learning goes next 01:00:16 - Inside NVIDIA's research org: "the mission is the boss" 01:04:03 - How NVIDIA decides who gets the GPUs 01:10:53 - Why NVIDIA still feels entrepreneurial after 33 years 01:12:58 - Why Bryan doesn't believe in the singularity 01:17:50 - The AI backlash 01:19:18 - The controversial case: open AI is safer than closed
1:22:57

Inside Nemotron and NVIDIA's AI lab: my conversation with Bryan Catanzaro (Bryan Catanzaro). NVIDIA is a chip company. So why does it put hundreds of researchers on building AI models - and then give them away for free? We go deep into the Nemotron models, what it takes to build a top AI lab, and the future of frontier AI. 01:33 - Is open source AI catching the frontier? 05:29 - Do closed labs blocking distillation slow open source down? 07:42 - Is the US falling behind China? 10:30 - Why companies actually choose open models 12:39 - A "crazy" 2008 bet: machine learning on GPUs 15:33 - Working with Andrew Ng and Dario Amodei at Baidu 17:41 - Coming back to NVIDIA: DLSS and the birth of Megatron 21:55 - The real reason NVIDIA builds its own models 24:28 - Is Moore's Law really dead? 33:37 - The Nemotron family: Nano, Super, Ultra 35:09 - Built for agents: why NVIDIA bets on speed 36:02 - How you train a 550B model in 4 bits 39:25 - Hybrid Mamba-Transformer, explained simply 42:31 - Mixture of experts, and why NVIDIA built NVL72 around it 47:26 - Why a 1-million-token context window matters 49:26 - Multi-token prediction: how the model predicts 5 tokens at once 52:47 - Multi-teacher distillation: teaching one model from many 58:01 - Where reinforcement learning goes next 01:00:16 - Inside NVIDIA's research org: "the mission is the boss" 01:04:03 - How NVIDIA decides who gets the GPUs 01:10:53 - Why NVIDIA still feels entrepreneurial after 33 years 01:12:58 - Why Bryan doesn't believe in the singularity 01:17:50 - The AI backlash 01:19:18 - The controversial case: open AI is safer than closed

Matt Turck

56,546 views • 22 days ago

DFlash makes Qwen 2.2x faster with no quality loss! We ran the same Qwen3.6-27B locally three ways on one RTX 6000: baseline, MTP, DFlash. The tasks only differ in one thing - how predictable the next word is: quicksort, describe a file in JSON, a logic puzzle, a sci-fi story. Outputs: Baseline: 44 tok/s · 1.00x MTP: 65 tok/s · 1.45x · 71% accepted DFlash: 98 tok/s · 2.20x · 30% accepted Baseline writes one token per step. MTP works inside the model itself and guesses 3 tokens ahead. DFlash is a separate small model that writes 15 tokens at once, and the big model only checks them. In JSON the same words repeat all the time, so most guesses were right: 152 tok/s, 3.4x speedup. In the story 9 guesses out of 10 were wrong. DFlash did all that extra work for nothing and became slower than baseline: 42 vs 44 tok/s. MTP guesses only 3 tokens, so a wrong guess costs very little: 46 tok/s and the win in that round. The output is identical in all three modes - DFlash is the pick for tasks with predictable output, like coding, and for chat and creative writing MTP works better. DFlash is now natively integrated into Atomic Chat on llama.cpp - speed up your Qwen models!
0:26

DFlash makes Qwen 2.2x faster with no quality loss! We ran the same Qwen3.6-27B locally three ways on one RTX 6000: baseline, MTP, DFlash. The tasks only differ in one thing - how predictable the next word is: quicksort, describe a file in JSON, a logic puzzle, a sci-fi story. Outputs: Baseline: 44 tok/s · 1.00x MTP: 65 tok/s · 1.45x · 71% accepted DFlash: 98 tok/s · 2.20x · 30% accepted Baseline writes one token per step. MTP works inside the model itself and guesses 3 tokens ahead. DFlash is a separate small model that writes 15 tokens at once, and the big model only checks them. In JSON the same words repeat all the time, so most guesses were right: 152 tok/s, 3.4x speedup. In the story 9 guesses out of 10 were wrong. DFlash did all that extra work for nothing and became slower than baseline: 42 vs 44 tok/s. MTP guesses only 3 tokens, so a wrong guess costs very little: 46 tok/s and the win in that round. The output is identical in all three modes - DFlash is the pick for tasks with predictable output, like coding, and for chat and creative writing MTP works better. DFlash is now natively integrated into Atomic Chat on llama.cpp - speed up your Qwen models!

atomic.chat

75,519 views • 11 days ago

Realtime-VLA FLASH tackles one of the biggest deployment bottlenecks for diffusion-based VLAs: inference latency. The key idea is speculative inference for flow-matching VLAs. A lightweight draft model predicts an action chunk, while the main model’s Action Expert verifies it in parallel using flow-consistency checks instead of running full denoising every replanning round. This lets the system replace many expensive 58 ms full inference rounds with speculative rounds as fast as 7.8 ms, reducing average latency to 19.1 ms and achieving a 3.04× speedup on LIBERO while largely preserving success rate. Interesting systems insight: they profile π0 and show VLM prefill is compute-bound, while Action Denoise is memory-bound. FLASH exploits this by reusing KV cache and parallelizing verification instead of repeatedly running sequential denoising.
1:54

Realtime-VLA FLASH tackles one of the biggest deployment bottlenecks for diffusion-based VLAs: inference latency. The key idea is speculative inference for flow-matching VLAs. A lightweight draft model predicts an action chunk, while the main model’s Action Expert verifies it in parallel using flow-consistency checks instead of running full denoising every replanning round. This lets the system replace many expensive 58 ms full inference rounds with speculative rounds as fast as 7.8 ms, reducing average latency to 19.1 ms and achieving a 3.04× speedup on LIBERO while largely preserving success rate. Interesting systems insight: they profile π0 and show VLM prefill is compute-bound, while Action Denoise is memory-bound. FLASH exploits this by reusing KV cache and parallelizing verification instead of repeatedly running sequential denoising.

Robots Digest 🤖

14,438 views • 2 months ago

watch this anon. i gave NVIDIA's biggest model ever a single task. 100 minutes and 440,000 tokens later, it had rendered nothing. not one important thing on the screen. this is Nemotron 3 Ultra. 550 billion parameters, a hybrid Mamba Transformer MoE, the largest model NVIDIA has ever shipped, and they built it specifically for long-running agentic coding. so i handed it exactly that: build a 3D scene from a spec, multiple files, iterate until the tests pass. the same task a frontier model one shotted in minutes. i genuinely wanted to be impressed. it ran for an hour and forty. burned through 440,000 tokens. wrote every file, passed its own tests, and proudly printed "task complete."the browser was blank. the 3D scene never rendered. not once. and the long horizon agentic behavior was genuinely good. it stayed on task the whole hour and forty, wrote real multi-file code, drove its own tools without derailing. it just couldn't turn any of that into something that actually runs. here's the part that gets me. it's a text model, it cannot see its own output. so it sat there looping on a broken vision tool, trying to "look" at the page, hitting error after error, never once reasoning its way out. it declared victory on an empty screen because it had no way to know the screen was empty. to be fair, i genuinely don't know what quant the NIM was serving, so maybe some of that's on the serving, not the model. but the biggest model NVIDIA has ever made, on the exact task it was designed for, couldn't tell it had built nothing in 100 minutes. same task on a local model, below thread👇.
4:07

watch this anon. i gave NVIDIA's biggest model ever a single task. 100 minutes and 440,000 tokens later, it had rendered nothing. not one important thing on the screen. this is Nemotron 3 Ultra. 550 billion parameters, a hybrid Mamba Transformer MoE, the largest model NVIDIA has ever shipped, and they built it specifically for long-running agentic coding. so i handed it exactly that: build a 3D scene from a spec, multiple files, iterate until the tests pass. the same task a frontier model one shotted in minutes. i genuinely wanted to be impressed. it ran for an hour and forty. burned through 440,000 tokens. wrote every file, passed its own tests, and proudly printed "task complete."the browser was blank. the 3D scene never rendered. not once. and the long horizon agentic behavior was genuinely good. it stayed on task the whole hour and forty, wrote real multi-file code, drove its own tools without derailing. it just couldn't turn any of that into something that actually runs. here's the part that gets me. it's a text model, it cannot see its own output. so it sat there looping on a broken vision tool, trying to "look" at the page, hitting error after error, never once reasoning its way out. it declared victory on an empty screen because it had no way to know the screen was empty. to be fair, i genuinely don't know what quant the NIM was serving, so maybe some of that's on the serving, not the model. but the biggest model NVIDIA has ever made, on the exact task it was designed for, couldn't tell it had built nothing in 100 minutes. same task on a local model, below thread👇.

Sudo su

32,589 views • 25 days ago

i'm running a 397 billion parameter model on a amd ai max box that sits on my desk and pulls less power than a gaming laptop. the model is Nex-N2-Pro, 397B-A17B, the open weight release people are putting next to gpt-5.5 on coding. i have it quantized to IQ1_M, 1.75 bits per weight, 90gb of weights loaded into the 128gb of unified memory on amd's strix halo igpu. watch the gpu in this recording. it spikes, it sustains, it does not fall over. that is the part the spec sheets never show you, not just that a 400b model loads, but that an integrated graphics chip holds the load and generates token after token, stable, no crash, no thermal cliff. and it is not a slideshow. roughly 18 tokens a second, faster than you can read. a frontier scale model producing usable output, fully local. no datacenter, no rented h100s, no api key, no permission. three years ago a model this size meant a server room and a budget to match. tonight it is a quiet box on my desk. this is the accessible tier almost nobody benchmarks honestly, and it is further along than the timeline thinks. the full breakdown is coming, rocm vs vulkan on this chip, and this little amd box head to head against the nvidia equivalent. stay tuned.
2:18

i'm running a 397 billion parameter model on a amd ai max box that sits on my desk and pulls less power than a gaming laptop. the model is Nex-N2-Pro, 397B-A17B, the open weight release people are putting next to gpt-5.5 on coding. i have it quantized to IQ1_M, 1.75 bits per weight, 90gb of weights loaded into the 128gb of unified memory on amd's strix halo igpu. watch the gpu in this recording. it spikes, it sustains, it does not fall over. that is the part the spec sheets never show you, not just that a 400b model loads, but that an integrated graphics chip holds the load and generates token after token, stable, no crash, no thermal cliff. and it is not a slideshow. roughly 18 tokens a second, faster than you can read. a frontier scale model producing usable output, fully local. no datacenter, no rented h100s, no api key, no permission. three years ago a model this size meant a server room and a budget to match. tonight it is a quiet box on my desk. this is the accessible tier almost nobody benchmarks honestly, and it is further along than the timeline thinks. the full breakdown is coming, rocm vs vulkan on this chip, and this little amd box head to head against the nvidia equivalent. stay tuned.

Sudo su

32,163 views • 1 month ago

got gemma 4 31B with MTP running on my DGX Spark. Hermes Agent did most of the legwork. baseline vs MTP on GB10: • c=1: 3.65 → 6.37 tok/s (1.74x) • c=4: 14.34 → 23.59 tok/s (1.65x) • c=8: 14.37 → 24.18 tok/s (1.68x) google says "up to 2x" — we're not quite there but it's real, not vapor. stack: DGX Spark / GB10 + gemma-4-31b-it + gemma-4-31b-it-assistant (MTP drafter) + vLLM built from PR 41745 MTP is basically a lightweight draft model that predicts multiple tokens while the big model verifies them all at once. smaller model does the busywork, bigger model just says yes/no. simple idea, weird to implement. next: tune the draft block size and see if we can push past 2x. also want to try it with Hermes Agent feeding prompts end to end. p.s: this was all done from telegram. Google DeepMind NVIDIA AI Developer
0:22

got gemma 4 31B with MTP running on my DGX Spark. Hermes Agent did most of the legwork. baseline vs MTP on GB10: • c=1: 3.65 → 6.37 tok/s (1.74x) • c=4: 14.34 → 23.59 tok/s (1.65x) • c=8: 14.37 → 24.18 tok/s (1.68x) google says "up to 2x" — we're not quite there but it's real, not vapor. stack: DGX Spark / GB10 + gemma-4-31b-it + gemma-4-31b-it-assistant (MTP drafter) + vLLM built from PR 41745 MTP is basically a lightweight draft model that predicts multiple tokens while the big model verifies them all at once. smaller model does the busywork, bigger model just says yes/no. simple idea, weird to implement. next: tune the draft block size and see if we can push past 2x. also want to try it with Hermes Agent feeding prompts end to end. p.s: this was all done from telegram. Google DeepMind NVIDIA AI Developer

Joey

23,088 views • 2 months ago

AMD might have disrupted Nvidia's entire cloud GPU rental business. In January at CES, AMD CEO Lisa Su demonstrated a $1,499 mini PC running the same class of AI model that currently costs companies $2,500 to $3,000 every month to rent from Nvidia-powered cloud servers. AMD's own branded version opened pre-orders this month at $3,999. Third party manufacturers have been selling the same chip since 2025 starting at $1,499. Here is exactly why this is dangerous for Nvidia. Nvidia's $75 billion quarterly revenue is built almost entirely on one business model, companies rent access to Nvidia GPUs through cloud providers like AWS and Lambda Labs to run AI. They pay monthly. Nvidia gets paid every time someone runs an AI model in the cloud. That recurring rental income is what turned Nvidia into a $5 trillion company. The AMD box eliminates that monthly fee permanently. One AI consultant switched from $2,800 per month in Nvidia cloud rental costs to $8 per month in electricity. The hardware paid for itself in 11 days. Over 8 months he generated $47,000 running the same AI workloads that previously left him paying Nvidia's ecosystem $2,800 every single month. Multiply that across thousands of enterprise customers and the revenue erosion becomes structural. Every business that buys this box stops paying cloud rental fees forever. Lawyers, doctors, banks, accountants, and financial advisors, businesses with sensitive data that cannot legally go to a cloud server represent billions in annual cloud GPU fees that Nvidia is now at risk of losing permanently. The threat is also closing in from the top. Google signed deals worth tens of billions with Anthropic and Meta to replace Nvidia with its own chips. Amazon built its own AI chips across AWS. Apple trained its AI on Google's chips, not Nvidia's. Custom silicon has grown from 21% of the AI chip market in 2025 to 28% in 2026. Nvidia's rental model only worked because serious AI compute had no alternative.
1:42

AMD might have disrupted Nvidia's entire cloud GPU rental business. In January at CES, AMD CEO Lisa Su demonstrated a $1,499 mini PC running the same class of AI model that currently costs companies $2,500 to $3,000 every month to rent from Nvidia-powered cloud servers. AMD's own branded version opened pre-orders this month at $3,999. Third party manufacturers have been selling the same chip since 2025 starting at $1,499. Here is exactly why this is dangerous for Nvidia. Nvidia's $75 billion quarterly revenue is built almost entirely on one business model, companies rent access to Nvidia GPUs through cloud providers like AWS and Lambda Labs to run AI. They pay monthly. Nvidia gets paid every time someone runs an AI model in the cloud. That recurring rental income is what turned Nvidia into a $5 trillion company. The AMD box eliminates that monthly fee permanently. One AI consultant switched from $2,800 per month in Nvidia cloud rental costs to $8 per month in electricity. The hardware paid for itself in 11 days. Over 8 months he generated $47,000 running the same AI workloads that previously left him paying Nvidia's ecosystem $2,800 every single month. Multiply that across thousands of enterprise customers and the revenue erosion becomes structural. Every business that buys this box stops paying cloud rental fees forever. Lawyers, doctors, banks, accountants, and financial advisors, businesses with sensitive data that cannot legally go to a cloud server represent billions in annual cloud GPU fees that Nvidia is now at risk of losing permanently. The threat is also closing in from the top. Google signed deals worth tens of billions with Anthropic and Meta to replace Nvidia with its own chips. Amazon built its own AI chips across AWS. Apple trained its AI on Google's chips, not Nvidia's. Custom silicon has grown from 21% of the AI chip market in 2025 to 28% in 2026. Nvidia's rental model only worked because serious AI compute had no alternative.

Bull Theory

26,668 views • 1 month ago

Etched came out of stealth at $800M and by lunch X had NVIDIA in the ground We do this every few months. A chip launches, the deck says killer, the timeline holds a funeral, and NVIDIA closes green anyway Etched hardwires the transformer into silicon. That is where the speed comes from, nearly the whole die on one job instead of the ~30% a GPU uses. It is also the trap. The day that chip tapes out is the best it will ever be. You cannot patch it. You burned progress into a wafer and now pray the field stops moving NVIDIA made the opposite bet. Same board, faster every quarter in software. Dynamo is pulling more tokens per watt out of the same rack, on version 1.0 The depreciation risk the bears aimed at NVIDIA for two years does not live at NVIDIA. It lives here, on the chip built to bury it Etched is not a fraud. It is a niche tool priced like a general one, and $800M is not enough to run a frontier supply chain. The rest get bought on the next down cycle Bury the lead, not the leader. Full case with Jack Farley and Max Wiethe on MTS And special thanks for Baseten for the cool T Shirt! Chapters 00:00 Switching from bonds to semis 00:33 What Etched actually is 01:31 Faster and cheaper, but how much HBM 02:56 Maturing market, not an NVIDIA killer 05:01 $1B in contracts and a Taiwan factory 05:20 Why these startups all get absorbed 06:48 Tiered inference and the obsolescence trap 09:39 Etched vs TPUs and Trainium 12:17 Is the CUDA moat weakening 13:21 Co-design, squeezing every token per watt 14:37 NVIDIA is a software company that sells a chip 14:58 Who is NVIDIA's most dangerous competitor 16:23 The NVIDIA killers, ranked 18:19 A rich man's game 18:43 AMD's MI500 vs Rubin Ultra 20:19 The neocloud business decision 22:46 Lightning round, Rambus the toll on HBM 25:11 The CXL run-up on Astera, Marvell, Credo 26:22 Use AI less, go to the booth 28:10 EDA is not dead
30:42

Etched came out of stealth at $800M and by lunch X had NVIDIA in the ground We do this every few months. A chip launches, the deck says killer, the timeline holds a funeral, and NVIDIA closes green anyway Etched hardwires the transformer into silicon. That is where the speed comes from, nearly the whole die on one job instead of the ~30% a GPU uses. It is also the trap. The day that chip tapes out is the best it will ever be. You cannot patch it. You burned progress into a wafer and now pray the field stops moving NVIDIA made the opposite bet. Same board, faster every quarter in software. Dynamo is pulling more tokens per watt out of the same rack, on version 1.0 The depreciation risk the bears aimed at NVIDIA for two years does not live at NVIDIA. It lives here, on the chip built to bury it Etched is not a fraud. It is a niche tool priced like a general one, and $800M is not enough to run a frontier supply chain. The rest get bought on the next down cycle Bury the lead, not the leader. Full case with Jack Farley and Max Wiethe on MTS And special thanks for Baseten for the cool T Shirt! Chapters 00:00 Switching from bonds to semis 00:33 What Etched actually is 01:31 Faster and cheaper, but how much HBM 02:56 Maturing market, not an NVIDIA killer 05:01 $1B in contracts and a Taiwan factory 05:20 Why these startups all get absorbed 06:48 Tiered inference and the obsolescence trap 09:39 Etched vs TPUs and Trainium 12:17 Is the CUDA moat weakening 13:21 Co-design, squeezing every token per watt 14:37 NVIDIA is a software company that sells a chip 14:58 Who is NVIDIA's most dangerous competitor 16:23 The NVIDIA killers, ranked 18:19 A rich man's game 18:43 AMD's MI500 vs Rubin Ultra 20:19 The neocloud business decision 22:46 Lightning round, Rambus the toll on HBM 25:11 The CXL run-up on Astera, Marvell, Credo 26:22 Use AI less, go to the booth 28:10 EDA is not dead

Ben Pouladian

29,969 views • 24 days ago

Micron is going to $4,000 and once you understand what inference actually is, the number stops sounding crazy (Save this). Dylan Patel just said that by 2030, OpenAI and Anthropic alone will need over 100 gigawatts of compute combined and by 2040, we may not even be measuring AI infrastructure in gigawatts anymore. We may be talking about terawatts. Every single one of those gigawatts needs memory to function. Without it, the compute is worthless. Most people heard that and thought about Nvidia but they should be thinking about Micron. Every AI model generating a response has two phases. The first is prefill, processing your prompt which is compute-heavy and the second is decode generating each word one token at a time and that phase is almost entirely memory-bound, not compute-bound. During decode, the GPU's processing units sit idle more than 95% of the time, waiting for data to arrive from memory. Google confirmed it in a research paper that decode-phase bottlenecks are dominated by memory bandwidth and capacity not raw compute. The GPU is not the bottleneck but the memory feeding the GPU is. This matters because inference is now where all the money lives. Training a model happens once, Inference happens billions of times a day every ChatGPT response, every Claude output, every agentic workflow running in the background and every one of those token streams is a billing event tied directly to memory performance. Adding more GPUs does not fix this because GPUs are already underutilized in inference because they are sitting idle waiting on memory. Adding more memory bandwidth and capacity is what directly reduces token cost, reduces latency, and allows the same cluster to serve dramatically more users simultaneously. Longer context windows compound the problem further, a model running a 1 million token context window requires dramatically more memory per session than a 10,000 token window, and every new model generation pushes context longer. The market treats memory as a downstream beneficiary of Nvidia orders. The correct framework is the opposite, Micron is the upstream constraint on how much value every Nvidia GPU can actually generate at inference scale. Micron guided Q4 to $50 billion in revenue, has HBM4 ramping at twice the pace of the prior generation, and CEO Sanjay Mehrotra has said supply will not catch demand before the end of 2027. At 8x forward earnings on $112 projected FY2027 EPS, Micron is the most undervalued infrastructure company in the entire AI stack. Inference is memory. Memory is Micron and the inference ramp has barely started. Milk Road Pro members are already up massively on this position and we're just getting started. If you want the full breakdown of what we're buying and why, come join us for just a dollar using the link below!
0:39

Micron is going to $4,000 and once you understand what inference actually is, the number stops sounding crazy (Save this). Dylan Patel just said that by 2030, OpenAI and Anthropic alone will need over 100 gigawatts of compute combined and by 2040, we may not even be measuring AI infrastructure in gigawatts anymore. We may be talking about terawatts. Every single one of those gigawatts needs memory to function. Without it, the compute is worthless. Most people heard that and thought about Nvidia but they should be thinking about Micron. Every AI model generating a response has two phases. The first is prefill, processing your prompt which is compute-heavy and the second is decode generating each word one token at a time and that phase is almost entirely memory-bound, not compute-bound. During decode, the GPU's processing units sit idle more than 95% of the time, waiting for data to arrive from memory. Google confirmed it in a research paper that decode-phase bottlenecks are dominated by memory bandwidth and capacity not raw compute. The GPU is not the bottleneck but the memory feeding the GPU is. This matters because inference is now where all the money lives. Training a model happens once, Inference happens billions of times a day every ChatGPT response, every Claude output, every agentic workflow running in the background and every one of those token streams is a billing event tied directly to memory performance. Adding more GPUs does not fix this because GPUs are already underutilized in inference because they are sitting idle waiting on memory. Adding more memory bandwidth and capacity is what directly reduces token cost, reduces latency, and allows the same cluster to serve dramatically more users simultaneously. Longer context windows compound the problem further, a model running a 1 million token context window requires dramatically more memory per session than a 10,000 token window, and every new model generation pushes context longer. The market treats memory as a downstream beneficiary of Nvidia orders. The correct framework is the opposite, Micron is the upstream constraint on how much value every Nvidia GPU can actually generate at inference scale. Micron guided Q4 to $50 billion in revenue, has HBM4 ramping at twice the pace of the prior generation, and CEO Sanjay Mehrotra has said supply will not catch demand before the end of 2027. At 8x forward earnings on $112 projected FY2027 EPS, Micron is the most undervalued infrastructure company in the entire AI stack. Inference is memory. Memory is Micron and the inference ramp has barely started. Milk Road Pro members are already up massively on this position and we're just getting started. If you want the full breakdown of what we're buying and why, come join us for just a dollar using the link below!

Milk Road AI

128,522 views • 23 days ago

NVIDIA MADE A $249 DESK CHIP THAT TURNS LOCAL LLMS INTO A TINY PRIVATE BACKEND jetson orin nano super is basically a pocket-size ai board: 67 TOPS, 1024-core ampere gpu, 8gb memory, 7-25w power draw. not a gaming pc, not a rented cloud box, just a small nvidia dev kit running llama, mistral, qwen and deepseek from your own table that changes the cost structure. the same automations people casually run through openai for $100-200/month can move the boring work local: summaries, drafts, code checks, document q&a, scraping cleanup, private agents. the board is $249 once, then around $2/month in electricity the setup is almost stupid. install ollama, pull a 7b model, aim your app at localhost. your workflow still talks to an openai-style api, but now the request never leaves the room and the bill does not grow every time a script loops no, it is not replacing claude for the hardest 20%. but it absolutely attacks the expensive background layer people forgot they were paying for every day cloud ai made people rent intelligence by the token. nvidia is quietly showing what happens when that meter sits under your desk instead.
4:08

NVIDIA MADE A $249 DESK CHIP THAT TURNS LOCAL LLMS INTO A TINY PRIVATE BACKEND jetson orin nano super is basically a pocket-size ai board: 67 TOPS, 1024-core ampere gpu, 8gb memory, 7-25w power draw. not a gaming pc, not a rented cloud box, just a small nvidia dev kit running llama, mistral, qwen and deepseek from your own table that changes the cost structure. the same automations people casually run through openai for $100-200/month can move the boring work local: summaries, drafts, code checks, document q&a, scraping cleanup, private agents. the board is $249 once, then around $2/month in electricity the setup is almost stupid. install ollama, pull a 7b model, aim your app at localhost. your workflow still talks to an openai-style api, but now the request never leaves the room and the bill does not grow every time a script loops no, it is not replacing claude for the hardest 20%. but it absolutely attacks the expensive background layer people forgot they were paying for every day cloud ai made people rent intelligence by the token. nvidia is quietly showing what happens when that meter sits under your desk instead.

Gipp 🦅

39,609 views • 1 month ago

Elon Musk just put a number on the flaw at the center of Nvidia’s empire. Wall Street has not done the math yet. Nvidia’s Blackwell is the most sought-after silicon on Earth. Every AI lab wants it. Every sovereign nation is bidding for it. Blackwell runs every model, for every company, in every data center on the planet. That universality built the empire. It is also the fracture point. Musk: “We believe the AI5 chip will be about a third of the power of an Nvidia Blackwell for roughly comparable performance. And much less than 10% of the cost.” One-third the power. Comparable performance. Less than ten percent of the cost. Musk: “This is a chip that is very much optimized for the Tesla AI software stack. It’s not meant to be a general purpose chip.” Nvidia builds silicon that serves a million different customers. Every transistor spent on universal compatibility is a transistor not dedicated to one task. Tesla is building silicon for exactly one customer. Itself. When you strip away every function you will never call, you do not get a lesser chip. You get a weapon. Here is what the market refuses to see. Data centers drink unlimited power from the grid. Robots run on batteries. Musk: “In order to have a functional robot, you have to have a great AI chip. And it needs to be an inexpensive chip and it needs to be very power efficient.” You cannot put a Blackwell inside a walking machine. It would drain the battery before it crossed the room. The entire AI revolution lives inside air-conditioned buildings bolted to the electrical grid. Musk is not competing for that market. He is engineering the silicon that survives outside of it. One-third the power is not a spec sheet footnote. It is the physics threshold that severs intelligence from the wall socket. Without that number, every robot on Earth stays tethered. With it, the algorithm walks. Less than ten percent of the cost is not a pricing strategy. It is the line where a machine brain stops being a capital expenditure and becomes a commodity component. When the chip inside a humanoid costs less than the motors in its legs, you do not manufacture hundreds of robots. You manufacture millions. Wall Street is valuing the AI revolution by who dominates the data center. Musk is building the only silicon designed to leave one. Nvidia built the brain of the cloud. Musk is building the brain of the physical world. No one has priced that in yet.
0:56

Elon Musk just put a number on the flaw at the center of Nvidia’s empire. Wall Street has not done the math yet. Nvidia’s Blackwell is the most sought-after silicon on Earth. Every AI lab wants it. Every sovereign nation is bidding for it. Blackwell runs every model, for every company, in every data center on the planet. That universality built the empire. It is also the fracture point. Musk: “We believe the AI5 chip will be about a third of the power of an Nvidia Blackwell for roughly comparable performance. And much less than 10% of the cost.” One-third the power. Comparable performance. Less than ten percent of the cost. Musk: “This is a chip that is very much optimized for the Tesla AI software stack. It’s not meant to be a general purpose chip.” Nvidia builds silicon that serves a million different customers. Every transistor spent on universal compatibility is a transistor not dedicated to one task. Tesla is building silicon for exactly one customer. Itself. When you strip away every function you will never call, you do not get a lesser chip. You get a weapon. Here is what the market refuses to see. Data centers drink unlimited power from the grid. Robots run on batteries. Musk: “In order to have a functional robot, you have to have a great AI chip. And it needs to be an inexpensive chip and it needs to be very power efficient.” You cannot put a Blackwell inside a walking machine. It would drain the battery before it crossed the room. The entire AI revolution lives inside air-conditioned buildings bolted to the electrical grid. Musk is not competing for that market. He is engineering the silicon that survives outside of it. One-third the power is not a spec sheet footnote. It is the physics threshold that severs intelligence from the wall socket. Without that number, every robot on Earth stays tethered. With it, the algorithm walks. Less than ten percent of the cost is not a pricing strategy. It is the line where a machine brain stops being a capital expenditure and becomes a commodity component. When the chip inside a humanoid costs less than the motors in its legs, you do not manufacture hundreds of robots. You manufacture millions. Wall Street is valuing the AI revolution by who dominates the data center. Musk is building the only silicon designed to leave one. Nvidia built the brain of the cloud. Musk is building the brain of the physical world. No one has priced that in yet.

Dustin

160,573 views • 3 months ago

A guy got tired of renting intelligence from the cloud, so he put a 120-billion-parameter model in his closet for $0 a month. One mini PC. The Minisforum MS-S1 MAX. 128GB of unified memory, and a trick that hands 96GB of it straight to the GPU. Here's how it breaks: That 96GB is the whole game. Enough to run GPT-OSS-120B fully local, and still keep the agent and dashboard on the same box. 120 billion parameters. 56.6 tokens a second. A 5,000-word essay in one pass at 110W and 68°C. Quiet. Cool. Local. Then Hermes Agent goes on top, pointed at localhost. No API key, because there's nothing to phone home to. It drafts, researches, runs tool calls, manages sub-agents. None of it leaves the network. The cloud version cost $10 to $20 a day, keys and customer data routed through machines he didn't own. Now it's a box in a closet. $0 a month. The chip is real. The build is brilliant. And somewhere in the first few minutes, the AI stopped being a bill he paid every morning and became a machine he owns, humming in a closet, working for free.
0:39

A guy got tired of renting intelligence from the cloud, so he put a 120-billion-parameter model in his closet for $0 a month. One mini PC. The Minisforum MS-S1 MAX. 128GB of unified memory, and a trick that hands 96GB of it straight to the GPU. Here's how it breaks: That 96GB is the whole game. Enough to run GPT-OSS-120B fully local, and still keep the agent and dashboard on the same box. 120 billion parameters. 56.6 tokens a second. A 5,000-word essay in one pass at 110W and 68°C. Quiet. Cool. Local. Then Hermes Agent goes on top, pointed at localhost. No API key, because there's nothing to phone home to. It drafts, researches, runs tool calls, manages sub-agents. None of it leaves the network. The cloud version cost $10 to $20 a day, keys and customer data routed through machines he didn't own. Now it's a box in a closet. $0 a month. The chip is real. The build is brilliant. And somewhere in the first few minutes, the AI stopped being a bill he paid every morning and became a machine he owns, humming in a closet, working for free.

shmidt

245,607 views • 1 month ago

SoftBank just sold its entire $5.83 billion Nvidia stake and if anything this move is actually a bull case for AI. Son needs $30 billion in cash this quarter to fund OpenAI, Ampere Computing, and Stargate infrastructure, plus a bunch of other AI bets. Nvidia was generating paper gains but zero liquidity. When you need that much cash deployed immediately, you liquidate positions that can move the needle.​ Son's not saying semiconductors are broken. He's saying the bigger returns are in the layer above the chips, the AI models, applications, and infrastructure that actually use Nvidia's GPUs. OpenAI, not Nvidia, is where he thinks the profit pool sits. That's a strategic bet on where value concentrates, not a bearish call on chip makers.​ This is actually a pattern. Son bought Nvidia in 2017 and sold in January 2019 (right before generative AI took off). People always say look at the gains he missed. But they ignore that he deployed that capital into PayPay, Coupang, and other companies. We see the same pattern here. He's trying to make the best use of his capital to make more bets and the portfolio returns validate that strategy.​ Son is also hedging his AI bet. Instead of staying concentrated in semiconductors, he's diversifying across the entire software and infrastructure layer. That's defensive positioning. It signals he's starting to think AI valuations might be a bit stretched, so he's spreading risk across more beneficiaries instead of leaning on one horse.​ Son's also optimizing for portfolio returns, not for holding the single best stock. He liquidates Nvidia to deploy capital into higher conviction bets where he thinks the actual value creation happens.
2:28

SoftBank just sold its entire $5.83 billion Nvidia stake and if anything this move is actually a bull case for AI. Son needs $30 billion in cash this quarter to fund OpenAI, Ampere Computing, and Stargate infrastructure, plus a bunch of other AI bets. Nvidia was generating paper gains but zero liquidity. When you need that much cash deployed immediately, you liquidate positions that can move the needle.​ Son's not saying semiconductors are broken. He's saying the bigger returns are in the layer above the chips, the AI models, applications, and infrastructure that actually use Nvidia's GPUs. OpenAI, not Nvidia, is where he thinks the profit pool sits. That's a strategic bet on where value concentrates, not a bearish call on chip makers.​ This is actually a pattern. Son bought Nvidia in 2017 and sold in January 2019 (right before generative AI took off). People always say look at the gains he missed. But they ignore that he deployed that capital into PayPay, Coupang, and other companies. We see the same pattern here. He's trying to make the best use of his capital to make more bets and the portfolio returns validate that strategy.​ Son is also hedging his AI bet. Instead of staying concentrated in semiconductors, he's diversifying across the entire software and infrastructure layer. That's defensive positioning. It signals he's starting to think AI valuations might be a bit stretched, so he's spreading risk across more beneficiaries instead of leaning on one horse.​ Son's also optimizing for portfolio returns, not for holding the single best stock. He liquidates Nvidia to deploy capital into higher conviction bets where he thinks the actual value creation happens.

StockMarket.News

317,998 views • 8 months ago