Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

Scale Python AI workloads from a single notebook to thousands of CPUs/GPUs. In #AzureFriday, Scott Hanselman 🌮 talks with Omar Shorbaji about building, training, and serving AI apps on Anyscale on Azure, powered by Ray, and running on AKS (without managing Kubernetes). Watch:

Azure's profile picture

Microsoft Azure

1,068,362 subscribers

16,550 Aufrufe • vor 16 Tagen •via X (Twitter)

#AzureFriday
LIVE

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

Building AI infrastructure at yotta-scale requires more than raw performance; it demands an open, modular rack design that can evolve across product generations, combining leadership compute engines with high-speed networking to connect thousands of accelerators into a single, unified system. Take a look at the Helios. Our rack-scale AI platform powered by MI455X GPUs and EPYC “Venice” CPUs and designed for trillion-parameter training.
0:39

Building AI infrastructure at yotta-scale requires more than raw performance; it demands an open, modular rack design that can evolve across product generations, combining leadership compute engines with high-speed networking to connect thousands of accelerators into a single, unified system. Take a look at the Helios. Our rack-scale AI platform powered by MI455X GPUs and EPYC “Venice” CPUs and designed for trillion-parameter training.

AMD

20,832 Aufrufe • vor 6 Monaten

🧐Build smarter AI workflows on Azure... Explore how Azure MCP Server and Azure Skills work together to extend AI capabilities across Azure Services—and see the full developer experience in VS Code from setup to interaction—in this #AzureFriday episode.
0:55

🧐Build smarter AI workflows on Azure... Explore how Azure MCP Server and Azure Skills work together to extend AI capabilities across Azure Services—and see the full developer experience in VS Code from setup to interaction—in this #AzureFriday episode.

Microsoft Azure

14,745 Aufrufe • vor 2 Monaten

Exciting to see the new RTX 50-series GPUs and Windows Subsystem for Linux powering models from Nvidia NIM and Azure AI Foundry on Windows 11 PCs. A game-changer for running AI on the edge.
1:39

Exciting to see the new RTX 50-series GPUs and Windows Subsystem for Linux powering models from Nvidia NIM and Azure AI Foundry on Windows 11 PCs. A game-changer for running AI on the edge.

Satya Nadella

234,980 Aufrufe • vor 1 Jahr

Dylan Patel: “CPUs are sold out.” $INTC earnings and conference call confirmed the looming CPU shortage. There are three main drivers: - Agentic systems, as CPUs do the planning and orchestration. - Increasing complexity of and demand for reinforcement learning environments that run on CPUs. - Increasing deployment of AI instilled apps. Models run on GPUs but apps themselves run on CPUs. Demand for all these will grow secularly as AI systems will only get more agentic, RL environments will only get more complex, and deployment will only accelerate. As a result, GPU-to-CPU ratio in AI workloads will move from 1:8 to 1:2 or maybe even 1:1 in the future. CPU supercycle is about to start. $AMD $INTC $ARM
1:44

Dylan Patel: “CPUs are sold out.” $INTC earnings and conference call confirmed the looming CPU shortage. There are three main drivers: - Agentic systems, as CPUs do the planning and orchestration. - Increasing complexity of and demand for reinforcement learning environments that run on CPUs. - Increasing deployment of AI instilled apps. Models run on GPUs but apps themselves run on CPUs. Demand for all these will grow secularly as AI systems will only get more agentic, RL environments will only get more complex, and deployment will only accelerate. As a result, GPU-to-CPU ratio in AI workloads will move from 1:8 to 1:2 or maybe even 1:1 in the future. CPU supercycle is about to start. $AMD $INTC $ARM

Oguz Erkan

93,419 Aufrufe • vor 2 Monaten

Today, we’re excited to announce and launch Reflex Cloud on Product Hunt! Reflex is an open-source framework for building and deploying data and AI web apps in pure Python. Frontend and Backend in Pure Python: No JavaScript required! With Reflex Cloud, you can now deploy, manage, and scale your Python apps with just a single command! If you're a Python developer, an upvote or a share would mean a lot to us :)💪❤️
3:04

Today, we’re excited to announce and launch Reflex Cloud on Product Hunt! Reflex is an open-source framework for building and deploying data and AI web apps in pure Python. Frontend and Backend in Pure Python: No JavaScript required! With Reflex Cloud, you can now deploy, manage, and scale your Python apps with just a single command! If you're a Python developer, an upvote or a share would mean a lot to us :)💪❤️

Reflex

55,024 Aufrufe • vor 1 Jahr

🔥 Launching the Fireworks AI Experimentation Platform (GA) and Build SDK (Beta)🔥 Top AI teams don't win by chance - they win by running 10X more experiments than competitors. Launching early, collecting feedback, tuning and selecting models, running offline evals, and validating impact with product metrics in a fast, continuous cycle is critical to build successful AI products. But, the need to manage complex infrastructure, wait for GPUs, reconcile differences between training and serving, and maintain reliability, slow AI teams down. Now, with the Firework’s experimentation platform, experimenting with open models just got a whole lot easier. With the help of our brand-new Python SDK (in beta), you can: → Stop waiting for GPUs, fine-tune and deploy 1000s of models on a single GPU deployment using Multi-LoRA → Programmatically run large-scale fine-tuning experiments → Scale evals with fully managed and autoscaling GPU deployments Check out our blog post to learn more and see a demo of running 125 experiments in 3 minutes on a single deployment:
4:41

🔥 Launching the Fireworks AI Experimentation Platform (GA) and Build SDK (Beta)🔥 Top AI teams don't win by chance - they win by running 10X more experiments than competitors. Launching early, collecting feedback, tuning and selecting models, running offline evals, and validating impact with product metrics in a fast, continuous cycle is critical to build successful AI products. But, the need to manage complex infrastructure, wait for GPUs, reconcile differences between training and serving, and maintain reliability, slow AI teams down. Now, with the Firework’s experimentation platform, experimenting with open models just got a whole lot easier. With the help of our brand-new Python SDK (in beta), you can: → Stop waiting for GPUs, fine-tune and deploy 1000s of models on a single GPU deployment using Multi-LoRA → Programmatically run large-scale fine-tuning experiments → Scale evals with fully managed and autoscaling GPU deployments Check out our blog post to learn more and see a demo of running 125 experiments in 3 minutes on a single deployment:

Fireworks AI

17,619 Aufrufe • vor 1 Jahr

Scott Hanselman joins the pod to talk dev insights from #MSBuild, the relationship between human agency and AI agents, and why curiosity and context are still key. Watch now: #AzurePodcast
0:25

Scott Hanselman joins the pod to talk dev insights from #MSBuild, the relationship between human agency and AI agents, and why curiosity and context are still key. Watch now: #AzurePodcast

Microsoft Azure

18,096 Aufrufe • vor 1 Jahr

Alibaba Cloud's Lingjun unifies HPN 7.0, AI Master Scheduler, and CPFS to make tens of thousands of GPUs work as one—slashing training time and hitting 93% utilization. Watch how we went from counting balls to cosmic-scale AI!
7:54

Alibaba Cloud's Lingjun unifies HPN 7.0, AI Master Scheduler, and CPFS to make tens of thousands of GPUs work as one—slashing training time and hitting 93% utilization. Watch how we went from counting balls to cosmic-scale AI!

Alibaba Cloud

228,081 Aufrufe • vor 5 Monaten

Unbiased AI -- at Scale Truly Open, inference at scale, running on Subnet 4. Powered by the Bittensor ecosystem. Full episode on Youtube and Spotify:
0:57

Unbiased AI -- at Scale Truly Open, inference at scale, running on Subnet 4. Powered by the Bittensor ecosystem. Full episode on Youtube and Spotify:

Openτensor Foundaτion

45,393 Aufrufe • vor 1 Jahr

At Google I/O, I sat down with Omar Sanseviero and 👩‍💻 Paige Bailey from Google DeepMind to talk about Gemma, open models, AI Studio, on-device AI, sovereign AI and the future of AI development. A great conversation on how building with AI is becoming more open, local and accessible.
26:51

At Google I/O, I sat down with Omar Sanseviero and 👩‍💻 Paige Bailey from Google DeepMind to talk about Gemma, open models, AI Studio, on-device AI, sovereign AI and the future of AI development. A great conversation on how building with AI is becoming more open, local and accessible.

Chubby♨️

13,741 Aufrufe • vor 1 Monat

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

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

Satya Nadella

907,397 Aufrufe • vor 8 Monaten

Now available: Sora in Azure OpenAI in Azure AI Foundry. This update empowers teams to: 🎨 Seamlessly integrate multimodal AI into apps and workflows via API-first design 🎨 Scale securely with built-in governance, identity, and billing Read more:
0:30

Now available: Sora in Azure OpenAI in Azure AI Foundry. This update empowers teams to: 🎨 Seamlessly integrate multimodal AI into apps and workflows via API-first design 🎨 Scale securely with built-in governance, identity, and billing Read more:

Microsoft Azure

66,524 Aufrufe • vor 1 Jahr

AI in the network is shifting the conversation beyond GPUs. At MWC 2026, Intel's Kevork Kechichian, EVP & GM of the Data Center Group, explains why inference workloads are bringing CPUs back into focus, especially in telco environments where power, space, and efficiency matter. With platforms like Intel Xeon and Clearwater Forest, operators are seeing up to 30% performance gains and 60% better power efficiency, helping scale AI workloads without overhauling infrastructure. As networks evolve toward AI-driven operations and 6G, one principle is becoming clear: The future of AI infrastructure is about using the right compute for the right workload.
1:14

AI in the network is shifting the conversation beyond GPUs. At MWC 2026, Intel's Kevork Kechichian, EVP & GM of the Data Center Group, explains why inference workloads are bringing CPUs back into focus, especially in telco environments where power, space, and efficiency matter. With platforms like Intel Xeon and Clearwater Forest, operators are seeing up to 30% performance gains and 60% better power efficiency, helping scale AI workloads without overhauling infrastructure. As networks evolve toward AI-driven operations and 6G, one principle is becoming clear: The future of AI infrastructure is about using the right compute for the right workload.

Six Five Media

131,501 Aufrufe • vor 4 Monaten

Our approach to AI infra is simple: build the most fungible and flexible fleet to meet the real world's needs across inference and training as Scott Guthrie shared with Alex Kantrowitz. And we are already doing it at scale today, as we power the biggest AI workloads like Copilot and ChatGPT, APIs that power 3P products & enterprise workloads and high scale training.
1:39

Our approach to AI infra is simple: build the most fungible and flexible fleet to meet the real world's needs across inference and training as Scott Guthrie shared with Alex Kantrowitz. And we are already doing it at scale today, as we power the biggest AI workloads like Copilot and ChatGPT, APIs that power 3P products & enterprise workloads and high scale training.

Satya Nadella

114,478 Aufrufe • vor 9 Monaten

We sat down with EvolutionIQOfficial's Tomas Vykruta to talk about AI in the insurance industry, importance of compelling UI, convincing the first customer to share training data with a startup, and other learnings from building a successful AI-powered business.
2:17

We sat down with EvolutionIQOfficial's Tomas Vykruta to talk about AI in the insurance industry, importance of compelling UI, convincing the first customer to share training data with a startup, and other learnings from building a successful AI-powered business.

Artem Burachenok, CEO @alloy.cx

102,816 Aufrufe • vor 3 Jahren

Announcing Cloudy, a platform that seamlessly handles your training infrastructure. You can rent a single H100 or a cluster of 1000 H100s, manage petabyte-scale storage volumes & seamlessly go from running experiments to managing large scale training runs on a single interface.
0:36

Announcing Cloudy, a platform that seamlessly handles your training infrastructure. You can rent a single H100 or a cluster of 1000 H100s, manage petabyte-scale storage volumes & seamlessly go from running experiments to managing large scale training runs on a single interface.

naklecha

33,138 Aufrufe • vor 9 Monaten

$AMD is becoming a two-engine AI story with EPYC CPUs for the growing compute needed around agentic workloads and Instinct GPUs for hyperscalers that need a credible alternative to $NVDA. The CPU side still looks underappreciated as AI agents require more high-core-count server compute and 2nm Venice should help AMD keep taking share as $INTC works through roadmap and manufacturing challenges. The bridge is Helios where AMD can package EPYC CPUs, Instinct GPUs and Pensando networking into a full rack-scale AI platform backed by real demand from $META, $ORCL and OpenAI.
0:59

$AMD is becoming a two-engine AI story with EPYC CPUs for the growing compute needed around agentic workloads and Instinct GPUs for hyperscalers that need a credible alternative to $NVDA. The CPU side still looks underappreciated as AI agents require more high-core-count server compute and 2nm Venice should help AMD keep taking share as $INTC works through roadmap and manufacturing challenges. The bridge is Helios where AMD can package EPYC CPUs, Instinct GPUs and Pensando networking into a full rack-scale AI platform backed by real demand from $META, $ORCL and OpenAI.

Shay Boloor

42,838 Aufrufe • vor 5 Tagen

“Any model, any agent, any accelerator, and any cloud.” Red Hat’s Brian Stevens explains why open AI infrastructure matters as enterprises scale agentic AI workloads, and how Red Hat and AMD are collaborating to help simplify deployment with AMD EPYC processors and AMD Instinct GPUs. Watch the full video on YouTube:
1:02

“Any model, any agent, any accelerator, and any cloud.” Red Hat’s Brian Stevens explains why open AI infrastructure matters as enterprises scale agentic AI workloads, and how Red Hat and AMD are collaborating to help simplify deployment with AMD EPYC processors and AMD Instinct GPUs. Watch the full video on YouTube:

AMD

17,208 Aufrufe • vor 2 Monaten

Had super cool discussion with Sam Altman on our strategy of creating the entire AI stack - GPUs, model, and apps. Willing to collaborate with India on all three.
0:57

Had super cool discussion with Sam Altman on our strategy of creating the entire AI stack - GPUs, model, and apps. Willing to collaborate with India on all three.

Ashwini Vaishnaw

82,670 Aufrufe • vor 1 Jahr