Загрузка видео...

Не удалось загрузить видео

Возникла проблема при загрузке этого видео. Это может быть связано с временными проблемами сети или видео может быть недоступно.

На главную

Modal engineer who teaches Full-Stack Deep Learning explained everything an AI engineer needs to know about GPUs in 20 minutes - better than $800 CUDA workshops. memory bandwidth -> SM occupancy -> kernel fusion -> arithmetic intensity -> know when you're compute-bound vs memory-bound. That loop is how you... show more

h100envy

2,239 subscribers

17,766 просмотров • 3 дней назад •via X (Twitter)

Наука и технологии

Anya Rossi• Live Now

Private livecam show

Комментарии: 0

Нет доступных комментариев

Здесь появятся комментарии из оригинального поста

Похожие видео

PyTorch core engineer at Meta turned CUDA kernel writing into a sport in 13 minutes - better than $1500 GPU programming bootcamps. profile the kernel -> find the bottleneck -> rewrite -> benchmark -> merge the winning code into PyTorch. That loop is how the open community now beats hand-tuned vendor kernels. GPU MODE community + KernelBot competition + winning kernel merged into the framework - that's the stack. Watch it, then steal the loop below.

h100envy

35,390 просмотров • 13 дней назад

Ex-Berkeley PhD who leads SGLang at xAI explained how they serve Grok on 100K GPUs in 23 minutes - better than $2000 inference-at-scale courses. split prefill and decode -> shard experts across GPUs -> route tokens per expert -> overlap comm and compute -> serve at DeepSeek-API-killing prices. That loop is why xAI runs Grok on SGLang and third parties beat DeepSeek's own API by 5x on cost. SGLang + prefill-decode disaggregation + expert parallelism + AMD MI300 - that's the stack. Watch and save it, then read the article below.

Ex-Berkeley PhD who leads SGLang at xAI explained how they serve Grok on 100K GPUs in 23 minutes - better than $2000 inference-at-scale courses. split prefill and decode -> shard experts across GPUs -> route tokens per expert -> overlap comm and compute -> serve at DeepSeek-API-killing prices. That loop is why xAI runs Grok on SGLang and third parties beat DeepSeek's own API by 5x on cost. SGLang + prefill-decode disaggregation + expert parallelism + AMD MI300 - that's the stack. Watch and save it, then read the article below.

h100envy

314,507 просмотров • 7 дней назад

Alibaba engineer who leads Qwen explained the future of open agent models in 25 minutes - better than $2000 LLM training courses. pre-train the base ->SFT -> RLHF -> tool use -> multi-modal -> ship a whole family (chat / VL / coder / math / QwQ). That loop is why Qwen quietly became the most downloaded open model family on Hugging Face. Qwen base + Qwen-VL + Qwen-Coder + QwQ reasoning - that's the stack. Watch and save it, then read the article below.

Alibaba engineer who leads Qwen explained the future of open agent models in 25 minutes - better than $2000 LLM training courses. pre-train the base ->SFT -> RLHF -> tool use -> multi-modal -> ship a whole family (chat / VL / coder / math / QwQ). That loop is why Qwen quietly became the most downloaded open model family on Hugging Face. Qwen base + Qwen-VL + Qwen-Coder + QwQ reasoning - that's the stack. Watch and save it, then read the article below.

h100envy

112,785 просмотров • 9 дней назад

Ex-NVIDIA engineer who built Unsloth explained RL, kernels, reasoning, quantization, and agents in 2 hours 42 minutes - better than $5000 fine-tuning bootcamps. pick the base model -> write triton kernels for 2x faster fine-tune -> quantize to 4-bit -> run GRPO/DPO -> ship a reasoning model on your single GPU. That loop is why Unsloth is the default way to fine-tune Llama, Qwen, Gemma, and Phi on hardware you already own. Unsloth + Triton kernels + 4-bit quantization + GRPO/DPO + single-GPU fine-tuning - that's the stack. Watch and save it, then fine-tune your first model tonight.

Ex-NVIDIA engineer who built Unsloth explained RL, kernels, reasoning, quantization, and agents in 2 hours 42 minutes - better than $5000 fine-tuning bootcamps. pick the base model -> write triton kernels for 2x faster fine-tune -> quantize to 4-bit -> run GRPO/DPO -> ship a reasoning model on your single GPU. That loop is why Unsloth is the default way to fine-tune Llama, Qwen, Gemma, and Phi on hardware you already own. Unsloth + Triton kernels + 4-bit quantization + GRPO/DPO + single-GPU fine-tuning - that's the stack. Watch and save it, then fine-tune your first model tonight.

h100envy

481,078 просмотров • 4 дней назад

Ex-Google engineer explained AI agent loops, harness, evals in 20 minutes - better than 500$ courses. trace every run → judge it with an LLM → diagnose → fix → ship. That loop is how agents self-improve over time. Agent loops + memory + harness + evals - thats the stack. Watch it, then save the framework below.

Ex-Google engineer explained AI agent loops, harness, evals in 20 minutes - better than 500$ courses. trace every run → judge it with an LLM → diagnose → fix → ship. That loop is how agents self-improve over time. Agent loops + memory + harness + evals - thats the stack. Watch it, then save the framework below.

Movez

553,889 просмотров • 15 дней назад

Ex-Google engineer reveals how to build AI agent loops, harnesses, LLM ops, and evals in 19 minutes. Trace → evaluate → diagnose → fix → ship → repeat. That loop is how agents self-improve over time. Agentic loops + harness + memory + evals - that’s the senior engineer stack. This is better than $500 paid courses on the same topic, explained in under 20 minutes. Watch it, then save the framework below.

Ex-Google engineer reveals how to build AI agent loops, harnesses, LLM ops, and evals in 19 minutes. Trace → evaluate → diagnose → fix → ship → repeat. That loop is how agents self-improve over time. Agentic loops + harness + memory + evals - that’s the senior engineer stack. This is better than $500 paid courses on the same topic, explained in under 20 minutes. Watch it, then save the framework below.

Codez

167,880 просмотров • 14 дней назад

CMU PhD who built the kernels NVIDIA now ships in TensorRT-LLM explained fast attention in 68 minutes - better than $1200 GPU programming courses. pick the attention pattern -> generate a fused CUDA kernel -> drop it into vLLM/SGLang -> same GPU, way more tokens per second. That loop is why FlashInfer now powers inference at NVIDIA, vLLM, SGLang, and half the serving stacks you use. FlashInfer + Triton + JIT-compiled kernels + paged-KV attention - that's the stack.

CMU PhD who built the kernels NVIDIA now ships in TensorRT-LLM explained fast attention in 68 minutes - better than $1200 GPU programming courses. pick the attention pattern -> generate a fused CUDA kernel -> drop it into vLLM/SGLang -> same GPU, way more tokens per second. That loop is why FlashInfer now powers inference at NVIDIA, vLLM, SGLang, and half the serving stacks you use. FlashInfer + Triton + JIT-compiled kernels + paged-KV attention - that's the stack.

h100envy

32,605 просмотров • 11 дней назад

The creator of High Bandwidth Memory (HBM) put a number on the AI build that should stop every infra investor cold. A cluster of a million GPUs runs at roughly 10-20% utilization (Save this). Kim Jung-ho spent thirty years building what feeds the GPU, and his claim is that the GPU is barely working. Here is what is actually happening. Every time a model generates output, the data has to be read out of memory, computed, and written back. The read and the write swallow almost the entire cycle. While that data moves, the GPU does nothing. It sits there, fully powered, fully paid for, waiting. By Kim's estimate the memory is doing only about 30 percent of the work it needs to do. The processor idles the rest. So a million installed GPUs run at 10 to 20 percent. You are not compute constrained. You are memory constrained, and the expensive part is standing around. Adding more GPUs does not fix this. It gives you more processors starving for the same data. Here is the part that decides the next decade. Memory can grow. When a cell cannot shrink any further, you stack it into a high-rise, layer on layer. A GPU cannot be stacked. It runs too hot and needs a cooler bolted to its back, so the one move that rescues memory is closed to the processor. The thing that can keep stacking compounds. The thing that cannot plateaus. The marginal dollar in an AI build now buys more by fixing the memory path than by bolting on another idle GPU. Which is why the companies that control memory bandwidth and supply are not suppliers to the AI trade. They are the AI trade.

The creator of High Bandwidth Memory (HBM) put a number on the AI build that should stop every infra investor cold. A cluster of a million GPUs runs at roughly 10-20% utilization (Save this). Kim Jung-ho spent thirty years building what feeds the GPU, and his claim is that the GPU is barely working. Here is what is actually happening. Every time a model generates output, the data has to be read out of memory, computed, and written back. The read and the write swallow almost the entire cycle. While that data moves, the GPU does nothing. It sits there, fully powered, fully paid for, waiting. By Kim's estimate the memory is doing only about 30 percent of the work it needs to do. The processor idles the rest. So a million installed GPUs run at 10 to 20 percent. You are not compute constrained. You are memory constrained, and the expensive part is standing around. Adding more GPUs does not fix this. It gives you more processors starving for the same data. Here is the part that decides the next decade. Memory can grow. When a cell cannot shrink any further, you stack it into a high-rise, layer on layer. A GPU cannot be stacked. It runs too hot and needs a cooler bolted to its back, so the one move that rescues memory is closed to the processor. The thing that can keep stacking compounds. The thing that cannot plateaus. The marginal dollar in an AI build now buys more by fixing the memory path than by bolting on another idle GPU. Which is why the companies that control memory bandwidth and supply are not suppliers to the AI trade. They are the AI trade.

Fireside Alpha

38,370 просмотров • 14 дней назад

Ex-vLLM core contributor explained how to make LLM inference 10x cheaper in 34 minutes - better than $3000 inference optimization bootcamps. request comes in -> check LMCache -> hit? load KV cache from CPU/SSD/remote -> skip prefill -> serve. That loop is why Bloomberg and other production stacks now push 300 terabytes of KV cache per week. LMCache + vLLM + CPU/SSD/remote storage + zero-copy CUDA kernels - that's the stack. Watch and save it, then wire the KV-offload into your inference stack.

Ex-vLLM core contributor explained how to make LLM inference 10x cheaper in 34 minutes - better than $3000 inference optimization bootcamps. request comes in -> check LMCache -> hit? load KV cache from CPU/SSD/remote -> skip prefill -> serve. That loop is why Bloomberg and other production stacks now push 300 terabytes of KV cache per week. LMCache + vLLM + CPU/SSD/remote storage + zero-copy CUDA kernels - that's the stack. Watch and save it, then wire the KV-offload into your inference stack.

h100envy

58,919 просмотров • 2 дней назад

Groq's LPU is faster than GPUs, handling requests and responding more quickly. Groq's LPUs don't need speedy data delivery like Nvidia GPUs do because they don't have HBM in their system. They use SRAM, which is about 20 times faster than what GPUs use. Since inference runs use way less data than model training, Groq's LPU is more energy-efficient. It reads less from external memory and uses less power than a Nvidia GPU for inference tasks. The LPU works differently from GPUs. It uses a Temporal Instruction Set Computer architecture, so it doesn't have to reload data from memory as often as GPUs do with High Bandwidth Memory (HBM). This helps avoid issues with HBM shortages and keeps costs down. If Groq's LPU is used in places that do AI processing, you might not need special storage for Nvidia GPUs. The LPU doesn't demand super-fast storage like GPUs do. Groq is claiming that its technology could replace GPUs in AI tasks with its powerful chip and software.

Groq's LPU is faster than GPUs, handling requests and responding more quickly. Groq's LPUs don't need speedy data delivery like Nvidia GPUs do because they don't have HBM in their system. They use SRAM, which is about 20 times faster than what GPUs use. Since inference runs use way less data than model training, Groq's LPU is more energy-efficient. It reads less from external memory and uses less power than a Nvidia GPU for inference tasks. The LPU works differently from GPUs. It uses a Temporal Instruction Set Computer architecture, so it doesn't have to reload data from memory as often as GPUs do with High Bandwidth Memory (HBM). This helps avoid issues with HBM shortages and keeps costs down. If Groq's LPU is used in places that do AI processing, you might not need special storage for Nvidia GPUs. The LPU doesn't demand super-fast storage like GPUs do. Groq is claiming that its technology could replace GPUs in AI tasks with its powerful chip and software.

Brian Roemmele

87,128 просмотров • 2 лет назад

Ex-Google engineer explains how to build memory for self-improving AI agents in 12 minutes. procedural (how to act / skills) + semantic (durable facts / profile) + episodic (dated events / chat history). Memory is an important part of an agentic loop. Memory + loops + harness + evals = self-improving agent system. This is better than $500 paid courses on the internet - in under 12 minutes. Watch it today, then save the detailed guide on building agent memory from scratch

Codez

87,612 просмотров • 13 дней назад

$NVDA hand-delivered the first Vera CPUs to Anthropic, OpenAI, SpaceX and Oracle Cloud. Vera is the company’s first custom CPU built for agentic AI marking its push deeper into the full compute stack as demand expands across CPUs, GPUs, networking and memory.

$NVDA hand-delivered the first Vera CPUs to Anthropic, OpenAI, SpaceX and Oracle Cloud. Vera is the company’s first custom CPU built for agentic AI marking its push deeper into the full compute stack as demand expands across CPUs, GPUs, networking and memory.

Shay Boloor

168,875 просмотров • 1 месяц назад

We're thrilled to share Modal Notebooks: a new, powerful cloud-hosted GPU notebook. It has modern real-time collaborative editing and is backed by our AI infrastructure — swap GPUs in seconds. Modal Notebooks are generally available, and you can start using them now. 🧵

We're thrilled to share Modal Notebooks: a new, powerful cloud-hosted GPU notebook. It has modern real-time collaborative editing and is backed by our AI infrastructure — swap GPUs in seconds. Modal Notebooks are generally available, and you can start using them now. 🧵

Eric Zhang

262,000 просмотров • 10 месяцев назад

GPU tradeoff series: A100 is not much more powerful than 4090 🫠 GPU Perf and Price: - 4090: 330 fp16 TFLOPs, $1,749 - A100 (80GB): 312 fp16 TFLOPs, $20,000 > A100 is 11.4X more pricy Training speed for GPT-2(124M) with llm.c: - 4090: 153K tokens/s - A100 (80GB): 195K tokens/s > A100 is only 1.3X faster (both trained using a single card, A100 llm.c training is shown in the video, 4090 video is in the quoted tweet) Conclusion: 4090 has a much better cost vs performance ratio Why: As in the H100 vs. 4090 comparison, the biggest difference between A100 and 4090 is their GPU memory size/bandwidth and cross-GPU communication bandwidth, which does not matter too much if your model can fit into a single 4090. Specs: 4090: - GPU memory size: 24GB - memory bandwidth: 1 TB/s - communication bandwidth: 64 GB/s A100: - GPU memory size: 80GB - memory bandwidth: 2 TB/s - communication bandwidth: 900 GB/s Nvidia killed off NVLink (a high-speed communication link that connects GPUs) on 4090. (Jensen Huang smiling face) If multiple 4090s could be interconnected via NVLink, their performance would be closer to datacenter-grade A100 GPUs, even for training larger models. Additionally, 4090 isn't allowed in datacenters, that's how Nvidia makes 💰💰💰

GPU tradeoff series: A100 is not much more powerful than 4090 🫠 GPU Perf and Price: - 4090: 330 fp16 TFLOPs, $1,749 - A100 (80GB): 312 fp16 TFLOPs, $20,000 > A100 is 11.4X more pricy Training speed for GPT-2(124M) with llm.c: - 4090: 153K tokens/s - A100 (80GB): 195K tokens/s > A100 is only 1.3X faster (both trained using a single card, A100 llm.c training is shown in the video, 4090 video is in the quoted tweet) Conclusion: 4090 has a much better cost vs performance ratio Why: As in the H100 vs. 4090 comparison, the biggest difference between A100 and 4090 is their GPU memory size/bandwidth and cross-GPU communication bandwidth, which does not matter too much if your model can fit into a single 4090. Specs: 4090: - GPU memory size: 24GB - memory bandwidth: 1 TB/s - communication bandwidth: 64 GB/s A100: - GPU memory size: 80GB - memory bandwidth: 2 TB/s - communication bandwidth: 900 GB/s Nvidia killed off NVLink (a high-speed communication link that connects GPUs) on 4090. (Jensen Huang smiling face) If multiple 4090s could be interconnected via NVLink, their performance would be closer to datacenter-grade A100 GPUs, even for training larger models. Additionally, 4090 isn't allowed in datacenters, that's how Nvidia makes 💰💰💰

Yuchen Jin

234,951 просмотров • 1 год назад

.Beff (e/acc) explains how thermodynamic computing fits into the current computing stack: "We don't do GPUs. They're called TSUs, thermodynamic sampling units. We think people are gonna put them next to GPUs or whatever their favorite accelerator is." "You don't need to run the whole workload on the TSU, but it could run certain parts of the workload, just like the GPU is complementary to the CPU initially." "The TSU is gonna help give you more performance per watt in your data center when combined with existing build-outs." "Let's say you're serving a large diffusion model. What we showed is that by using a TSU, you can use something called an energy-based model instead of a simple probabilistic kernel." "Overall you can have basically 100 times the performance per watt of just running on pure GPUs because our chips are much lower wattage."

.Beff (e/acc) explains how thermodynamic computing fits into the current computing stack: "We don't do GPUs. They're called TSUs, thermodynamic sampling units. We think people are gonna put them next to GPUs or whatever their favorite accelerator is." "You don't need to run the whole workload on the TSU, but it could run certain parts of the workload, just like the GPU is complementary to the CPU initially." "The TSU is gonna help give you more performance per watt in your data center when combined with existing build-outs." "Let's say you're serving a large diffusion model. What we showed is that by using a TSU, you can use something called an energy-based model instead of a simple probabilistic kernel." "Overall you can have basically 100 times the performance per watt of just running on pure GPUs because our chips are much lower wattage."

MTS

42,570 просмотров • 1 месяц назад

This is my favorite clip of the new Elon pod. He opens up saying xAI struggles with memory usage/bandwidth and CUDA kernel optimization (matmul, attention, MoE, etc). If you are good kernel or performance engineering in general, you should apply. Steer the world in a better direction.

This is my favorite clip of the new Elon pod. He opens up saying xAI struggles with memory usage/bandwidth and CUDA kernel optimization (matmul, attention, MoE, etc). If you are good kernel or performance engineering in general, you should apply. Steer the world in a better direction.

Elliot Arledge

163,922 просмотров • 6 месяцев назад

Everyone’s still talking about GPUs. Meanwhile, memory is now the real constraint for AI at scale. From GTC, Patrick Moorhead & Daniel Newman are joined by Paul Cho & Indong Kim from Samsung Semiconductor US to discuss why bandwidth, not compute, is starting to define AI performance.

Everyone’s still talking about GPUs. Meanwhile, memory is now the real constraint for AI at scale. From GTC, Patrick Moorhead & Daniel Newman are joined by Paul Cho & Indong Kim from Samsung Semiconductor US to discuss why bandwidth, not compute, is starting to define AI performance.

Six Five Media

56,512 просмотров • 3 месяцев назад

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!

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

128,079 просмотров • 13 дней назад

SITUATION EXPLAINED: Cerebras raised $5.55 billion in their IPO and closing their first day of trading valued at $66 billion, making it the biggest US tech IPO since Snowflake in 2020. Cerebras makes Wafer-Scale Engine chips built for AI inference. We asked Sarah Fong the main difference between wafer-scale chips and traditional GPUs: - GPUs are great at parallel work (graphics, training) - AI inference is sequential, AKA one token at a time This causes the "memory wall" problem: - Every GPU core needs model weights, KV cache, and activations to do its math - On a GPU, that data lives in off-chip memory (HBM) - Cores constantly load and offload from off-chip memory, which is a huge bottleneck; hardware accounts for ~70% of inference latency Cerebras' chips: -Dinner-plate sized (vs. GPUs which are palm-sized) with tens of thousands of cores -Memory sits directly on top of the cores as distributed SRAM -Weights and KV cache can be accessed at on-chip speeds in the PB/s range, compared with off-chip speeds in the TB/s range achieved by GPUs with HBM.

SITUATION EXPLAINED: Cerebras raised $5.55 billion in their IPO and closing their first day of trading valued at $66 billion, making it the biggest US tech IPO since Snowflake in 2020. Cerebras makes Wafer-Scale Engine chips built for AI inference. We asked Sarah Fong the main difference between wafer-scale chips and traditional GPUs: - GPUs are great at parallel work (graphics, training) - AI inference is sequential, AKA one token at a time This causes the "memory wall" problem: - Every GPU core needs model weights, KV cache, and activations to do its math - On a GPU, that data lives in off-chip memory (HBM) - Cores constantly load and offload from off-chip memory, which is a huge bottleneck; hardware accounts for ~70% of inference latency Cerebras' chips: -Dinner-plate sized (vs. GPUs which are palm-sized) with tens of thousands of cores -Memory sits directly on top of the cores as distributed SRAM -Weights and KV cache can be accessed at on-chip speeds in the PB/s range, compared with off-chip speeds in the TB/s range achieved by GPUs with HBM.

MTS

44,057 просмотров • 1 месяц назад

Quai Network's Iron Age Testnet has seen more than 3,000 GPUs come online to mine! If you're using GPUs to #MineQuai, be sure to share a picture in the Quai Discord 📸

Quai Network's Iron Age Testnet has seen more than 3,000 GPUs come online to mine! If you're using GPUs to #MineQuai, be sure to share a picture in the Quai Discord 📸

Quai Network ⚡️💵

27,599 просмотров • 2 лет назад