正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

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

16,540 subscribers

33,343 次观看 • 11 个月前 •via X (Twitter)

科学技术教育

Anya Rossi• Live Now

Private livecam show

45 条评论

naklecha 的头像

naklecha11 个月前

In the next few years, we are going to see an explosion of teams training models, researching ideas or running experiments on GPUs. The GPU shortage problem will get worse, while the number of teams renting compute will increase. Cloudy is betting on this :)

naklecha 的头像

naklecha11 个月前

Try the beta: Single node access & network volumes are available to everyone. For multi-node, we are doing a rolling release, one team at a time.

atharva ☆ 的头像

atharva ☆11 个月前

the age of amazing infra for independent researchers is upon us

naklecha 的头像

naklecha11 个月前

real and true

sahir 的头像

sahir11 个月前

tuff ❤️‍🔥

naklecha 的头像

naklecha11 个月前

ty king

kanishk 的头像

kanishk11 个月前

good stuff

naklecha 的头像

naklecha11 个月前

thank you!! just getting started :’)

KShivendu 🌁 的头像

KShivendu 🌁11 个月前

Congrats on the launch! Feels similar to modal. Curious to know how do you plan to differentiate? :)

naklecha 的头像

naklecha11 个月前

cloudy gives users full access to their compute & building tooling on top. personally i prefer this & it makes running experiments & long training runs a lot cheaper.

λux 的头像

λux11 个月前

⚡️⚡️⚡️

naklecha 的头像

naklecha11 个月前

:’)

Florian Brand11 个月前

BANGER 🔥

naklecha 的头像

naklecha11 个月前

AND WE ARE JUST GETTING STARTED 🔥🔥🔥

Braeden Norris 的头像

Braeden Norris11 个月前

Looks sick!

naklecha 的头像

naklecha11 个月前

thank you!!

vineet 的头像

vineet11 个月前

Yo congrats on re launch

naklecha 的头像

naklecha11 个月前

<3 ty vineet

thegodfather 的头像

thegodfather10 个月前

congrats on the launch, really like the UI !

naklecha 的头像

naklecha10 个月前

thank you! working on polishing up the ui and letting you edit files directly from the browser this week. it’s going to get a lot better!

thegodfather 的头像

thegodfather10 个月前

it already looks amazing, great work ! i can recognize some of the tools you used but was there any inspiration you based it from ? UI is a skill i have yet to master lol

naklecha 的头像

naklecha10 个月前

@shadcn’s library is incredible

varepsilon 的头像

varepsilon11 个月前

absolutely cooked on the ui

naklecha 的头像

naklecha11 个月前

thanks for reviewing it last week 😋

✨Rodmar パブロ✨ 的头像

✨Rodmar パブロ✨11 个月前

let's goooo!!

naklecha 的头像

naklecha11 个月前

YES ❤️

javi 的头像

javi11 个月前

goated

naklecha 的头像

naklecha11 个月前

omg hii javi

tanisha ✨ 的头像

tanisha ✨11 个月前

🔥🔥🔥

naklecha 的头像

naklecha11 个月前

tyyy

naveen 的头像

naveen11 个月前

Congrats on the revamp! Excited to try out the new site. Just a question - are you gonna migrate any existing credits from the old site?

naklecha 的头像

naklecha11 个月前

yess

naveen 的头像

naveen11 个月前

Sweet!

Josh 的头像

Josh11 个月前

Definitely will look into this. Need to up my game here 👌🤝

arya ☁️ 的头像

arya ☁️11 个月前

this is amazing can’t wait to try!!

naklecha 的头像

naklecha11 个月前

lemme know how it goes

Eigenron 的头像

Eigenron11 个月前

love the new ui

naklecha 的头像

naklecha11 个月前

tyty <3

Saïd Aitmbarek 的头像

Saïd Aitmbarek11 个月前

based concept & platform observing more & more AI native startups bullish on platforms for gpu infra + fine-tuning models let's launch/spotlight you guys on when it helps btw

Gurung | floodreliefnepal.com 的头像

Gurung | floodreliefnepal.com11 个月前

congrats on the launch nikhil :)

ßadroo Dz 的头像

ßadroo Dz8 个月前

bitcoin continues to grow due to strong global demand.

KShivendu 🌁 的头像

KShivendu 🌁11 个月前

Congrats on the launch! Reminder me of modal except it probably gives full access to dedicated machines instead of serverless stuff :D

naklecha 的头像

naklecha11 个月前

yep

KShivendu 🌁 的头像

KShivendu 🌁11 个月前

You replied very quickly. I was about to delete my prev comment xD

Baasanjaw Baasanjaw 的头像

Baasanjaw Baasanjaw8 个月前

Reliable resources for 5G technology include reputable textbooks, peer-reviewed articles, and official documentation.

相关视频

today, i'm excited to release cloudy -- a platform which enables you to sync & mount storage volumes on gpus sourced from popular cloud providers like lambdalabs, runpod, hyperbolic, nebius etc. save your team countless hours every month.

today, i'm excited to release cloudy -- a platform which enables you to sync & mount storage volumes on gpus sourced from popular cloud providers like lambdalabs, runpod, hyperbolic, nebius etc. save your team countless hours every month.

naklecha

74,100 次观看 • 1 年前

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:

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:

Microsoft Azure

16,550 次观看 • 2 个月前

Introducing Lab: A full-stack platform for training your own agentic models Build, evaluate and train on your own environments at scale without managing the underlying infrastructure. Giving everyone their own frontier AI lab.

Introducing Lab: A full-stack platform for training your own agentic models Build, evaluate and train on your own environments at scale without managing the underlying infrastructure. Giving everyone their own frontier AI lab.

Prime Intellect

3,155,191 次观看 • 7 个月前

We believe that intelligence should not arrive preconfigured. Together AI is now available directly inside the Adaption platform, connecting Adaptive Data with large-scale training in a single workflow. One platform, end to end. Stop inheriting intelligence. Shape it.

We believe that intelligence should not arrive preconfigured. Together AI is now available directly inside the Adaption platform, connecting Adaptive Data with large-scale training in a single workflow. One platform, end to end. Stop inheriting intelligence. Shape it.

adaption

25,869 次观看 • 4 个月前

Excited to release 🚀NeRF-XL🚀, a principled method to distribute your NeRF's parameters to multiple GPUs! It is heuristics-free and mathematically equiv. with training your NeRF on a single (large) GPU. Let's scale it up!🌎

Excited to release 🚀NeRF-XL🚀, a principled method to distribute your NeRF's parameters to multiple GPUs! It is heuristics-free and mathematically equiv. with training your NeRF on a single (large) GPU. Let's scale it up!🌎

Ruilong Li

55,288 次观看 • 2 年前

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.

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 次观看 • 8 个月前

HERMES Agent just launched a desktop app for managing autonomous AI agents at scale You can now: - Run multiple agents from one interface - Maintain persistent memory - Automate long-running workflows - Keep agents operating continuously

HERMES Agent just launched a desktop app for managing autonomous AI agents at scale You can now: - Run multiple agents from one interface - Maintain persistent memory - Automate long-running workflows - Keep agents operating continuously

Vaishnavi

58,365 次观看 • 3 个月前

When people talk about robotics, they usually talk about models, data, or hardware. Few people talk about the infrastructure that lets you iterate on all three quickly. Today we're publishing how we trained Dyna-2 on over 1,000,000 hours of egocentric video, repeatably. At this scale, most of what worked at ten thousand hours did not hold up: • ingestion throughput was capped at 14,000 episode-hours per week — a million hours would have taken over a year • building a training manifest took 48 hours before a run could even start • reading a petabyte from cloud storage during training left GPUs exposed to latency and packet loss 🧵

When people talk about robotics, they usually talk about models, data, or hardware. Few people talk about the infrastructure that lets you iterate on all three quickly. Today we're publishing how we trained Dyna-2 on over 1,000,000 hours of egocentric video, repeatably. At this scale, most of what worked at ten thousand hours did not hold up: • ingestion throughput was capped at 14,000 episode-hours per week — a million hours would have taken over a year • building a training manifest took 48 hours before a run could even start • reading a petabyte from cloud storage during training left GPUs exposed to latency and packet loss 🧵

Dyna Robotics

162,753 次观看 • 27 天前

Just shipped scale orders on Pacifica Scale orders are limit orders that allow you to place multiple limit orders in a predetermined range. We made it a bit different: 1. You can drag the price ranges directly on the chart. 2. You can visualize the orders and avg price for that scale order before the orders are placed. 3. We also made scale orders very easy to manage. If you want to cancel a particular scale order - instead of having to cancel them one by one, you can click on open orders -> scale orders to cancel a particular scale order.

Just shipped scale orders on Pacifica Scale orders are limit orders that allow you to place multiple limit orders in a predetermined range. We made it a bit different: 1. You can drag the price ranges directly on the chart. 2. You can visualize the orders and avg price for that scale order before the orders are placed. 3. We also made scale orders very easy to manage. If you want to cancel a particular scale order - instead of having to cancel them one by one, you can click on open orders -> scale orders to cancel a particular scale order.

Jose

23,512 次观看 • 8 个月前

If your supplier can’t handle your peak volume, you don’t have a scalable business—you have a bottleneck. Dropped by one of our city warehouses today. That massive truck in the back? That is a single order for just one of our clients. With 11 global fulfillment centers operating at this scale, we build the infrastructure so you can focus entirely on marketing. Let's scale your store to the next level. Join us at CJ Dropshipping.

If your supplier can’t handle your peak volume, you don’t have a scalable business—you have a bottleneck. Dropped by one of our city warehouses today. That massive truck in the back? That is a single order for just one of our clients. With 11 global fulfillment centers operating at this scale, we build the infrastructure so you can focus entirely on marketing. Let's scale your store to the next level. Join us at CJ Dropshipping.

Andy Chou

3,448,475 次观看 • 3 个月前

Scale your generations with BudgetPixel API & MCP. Access a large selection of image, video, music, and SFX models through one platform, then use them for bulk generation workflows without managing separate integrations.

Scale your generations with BudgetPixel API & MCP. Access a large selection of image, video, music, and SFX models through one platform, then use them for bulk generation workflows without managing separate integrations.

BudgetPixel AI

22,829 次观看 • 5 天前

Facing the crucible of JRTC. 10th Mountain Div. Soldiers participate in a Joint Readiness Training Center rotation at Fort Polk, La. JRTC is a combat training center that provides realistic deployment training scenarios in simulated large-scale combat operations to build readiness to support globally deployable missions.

Facing the crucible of JRTC. 10th Mountain Div. Soldiers participate in a Joint Readiness Training Center rotation at Fort Polk, La. JRTC is a combat training center that provides realistic deployment training scenarios in simulated large-scale combat operations to build readiness to support globally deployable missions.

U.S. Army

107,980 次观看 • 11 个月前

🤖 Robotics often faces a chicken and egg problem: no web-scale robot data for training (unlike CV or NLP) b/c robots aren't deployed yet & vice-versa. Introducing VRB: Use large-scale human videos to train a *general-purpose* affordance model to jumpstart any robotics paradigm!

🤖 Robotics often faces a chicken and egg problem: no web-scale robot data for training (unlike CV or NLP) b/c robots aren't deployed yet & vice-versa. Introducing VRB: Use large-scale human videos to train a general-purpose affordance model to jumpstart any robotics paradigm!

Deepak Pathak

70,712 次观看 • 3 年前

We're building Roblox to be a platform where anyone, anywhere, can be a creator. Generative AI is part of that vision: tools that help you bring your ideas to life seamlessly and interactively. More in our blog post:

We're building Roblox to be a platform where anyone, anywhere, can be a creator. Generative AI is part of that vision: tools that help you bring your ideas to life seamlessly and interactively. More in our blog post:

Roblox

490,755 次观看 • 3 年前