正在加载视频...

视频加载失败

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. 🧵

262,757 次观看 • 1 年前 •via X (Twitter)

46 条评论

Eric Zhang 的头像
Eric Zhang1 年前

I started hacking on Jupyter notebooks because they’re incredibly useful, a pillar of research and fast iteration. But it’s too easy to lose your work, the documents are siloed, and few providers offer high-end GPUs. That’s not good DevEx!

Eric Zhang 的头像
Eric Zhang1 年前

So, we reimagined notebooks from the ground up: - Each notebook can have up to 8 B200 GPUs and starts in seconds - Use arbitrary custom images - Modern features: AI, LSP, semantic highlighting - Jupyter Widgets synced across users - A beautiful, sleek UI

Eric Zhang 的头像
Eric Zhang1 年前

During our short beta period last month, thousands of developers have already started using Modal Notebooks, and they ran 200,000+ code cells. One of those is @SunoMusic, where researchers use notebooks to share models and experiments across the company.

Eric Zhang 的头像
Eric Zhang1 年前

More info in our product announcement: Modal Notebooks are part of our goal to build a cloud platform better tailored for developers. We hope they help people to work together and ultimately bring simpler, better technology into the world.

Teknium 🪽 的头像
Teknium 🪽1 年前

This is cozy for researchers imo

Jonathan Beckman 的头像
Jonathan Beckman1 年前

@marimo_io notebooks next please 🙏

Eric Zhang 的头像
Eric Zhang1 年前

@marimo_io Would love to see it!

Khart Nicksome 的头像
Khart Nicksome1 年前

yikes man give me better code and devops tools not gay notebooks

Jeff Huber 的头像
Jeff Huber1 年前

finally!

EDITH 的头像
EDITH1 年前

LFG!!! finally an alternative to colab

Amir Rustamzadeh 的头像
Amir Rustamzadeh1 年前

👌migrating Colab projects to Modal.

Elliot Arledge 的头像
Elliot Arledge1 年前

super! can you do a demo soon on compiling and running a vecadd cuda kernel? lots of people trying to learn cuda and this might just be the best tool for it!

Eric Zhang 的头像
Eric Zhang1 年前

It should work! nvcc is built into the notebook base image – want to try making one and sharing the link?

Elliot Arledge 的头像
Elliot Arledge1 年前

nvcc is, but not the drivers.

Eric Zhang 的头像
Eric Zhang1 年前

Got it, okay we'll have to fix that!

Surya 的头像
Surya1 年前

I thought you guys are partnering with @marimo_io

Eric Zhang 的头像
Eric Zhang1 年前

@suryatkin @marimo_io We are! Modal Sandboxes also power Marimo's cloud offering, and actually I showed them what we were working on to encourage them to try using Sandboxes :) Big fan of the innovative work they are doing

Surya 的头像
Surya1 年前

@marimo_io Awesome

lucas gelfond 的头像
lucas gelfond1 年前

yoooooooo

Eric Zhang 的头像
Eric Zhang1 年前

🥰 coincides with your move to nyc!

lucas gelfond 的头像
lucas gelfond1 年前

yes!!

Aaliya 的头像
Aaliya1 年前

This looks awesome @ekzhang1 Real-time collab + instant GPU swaps is a huge time saver

Saïd Aitmbarek 的头像
Saïd Aitmbarek1 年前

so bullish on more gpu/notebook infra for builders let's launch modal on when it helps guys

Cathy Zhou 的头像
Cathy Zhou1 年前

So excited to try this 🤯

Tobe Duru 的头像
Tobe Duru1 年前

game changer for data teams

topmass 的头像
topmass1 年前

this looks awesome, I want to work with and tinker !

tian/天 的头像
tian/天1 年前

Incredible👏

Mihir 的头像
Mihir1 年前

oh this is super helpful! any plans to having a cli client to run .cu scripts?

Antoine Weber 的头像
Antoine Weber1 年前

looks cool and great presentation

raku 的头像
raku1 年前

awesome!

Himanshu Kumar 的头像
Himanshu Kumar1 年前

Collaboration and powerful GPUs, a compelling combination. This could significantly accelerate development cycles. I wonder what new possibilities this unlocks.

Arvind 的头像
Arvind1 年前

Thanks man 😊🙏

Surya Dantuluri 的头像
Surya Dantuluri1 年前

been using modal notebooks for the past few weeks to run experiments. huge

oshaks 的头像
oshaks1 年前

YESS THANK YOU🤦🏿‍♂️😭😭😭😭😭

NotSuj 的头像
NotSuj1 年前

Lfg

Agam Chaudhary 的头像
Agam Chaudhary1 年前

Modal Notebooks sound like a game-changer for anyone needing flexible, powerful cloud-based GPU resources! The real-time collaboration and seamless GPU swapping are huge wins for productivity. How do you see this enhancing workflows for AI research or development teams?

vish 的头像
vish1 年前

congratulations!

Brandon Keyyy 的头像
Brandon Keyyy1 年前

Modal's making cloud-hosted slick with fast GPUs and teamwork magic! It's like giving every AI project an energy drink. Ready to see ideas launch into the stratosphere!

David Gwyer - AI & ML Researcher/Engineer 的头像
David Gwyer - AI & ML Researcher/Engineer1 年前

Looks great Eric! One thing that cropped up early in our fastai course (by @answerdotai) study group was a robust way to collaborate live on a notebook (e.g. like you can for Google docs). Is this possible with Modal notebooks?

Patrick Emedom-Nnamdi 的头像
Patrick Emedom-Nnamdi1 年前

This is awesome @ekzhang1 !

August Sha 的头像
August Sha1 年前

why is this better than google colab? don't they also have fast startup GPUs and even a free one?

⟁ndrew V 的头像
⟁ndrew V1 年前

Awesome, very awesome!

browncapitalist 的头像
browncapitalist1 年前

Nice

Vinit Wadgaonkar 的头像
Vinit Wadgaonkar1 年前

Congrats to the Modal team on launching Modal Notebooks 🎉 The real-time collaboration + GPU flexibility looks incredible. As a recent M.S. grad working on LLMs/AI infra (currently on OPT), I’d be thrilled to explore opportunities to contribute at Modal — is your team hiring?

Welcom.BTB 的头像
Welcom.BTB11 个月前

互关,回关下,谢谢

Fredrik Cassel 的头像
Fredrik Cassel1 年前

so strong 🚀

相关视频

Massive breakthrough here! Someone fixed every major flaw in Jupyter Notebooks. The .ipynb format is stuck in 2014. It was built for a different era - no cloud collaboration, no AI agents, no team workflows. Change one cell, and you get 50+ lines of JSON metadata in your git diff. Code reviews become a nightmare. Want to share a database connection across notebooks? Configure it separately in each one. Need comments or permissions? Too bad. Jupyter works for solo analysis but breaks for teams building production AI systems. Deepnote just open-sourced the solution (Apache 2.0 license) They've built a new notebook standard that actually fits modern workflows: ↳ Human-readable YAML - Git diffs show actual code changes, not JSON noise. Code reviews finally work. ↳ Project-based structure - Multiple notebooks share integrations, secrets, and environment settings. Configure once, use everywhere. ↳ 23 new block - SQL, interactive inputs, charts, and KPIs as first-class citizens. Build data apps, not just analytics notebooks. ↳ Multi-language support - Python and SQL in one notebook. Modern data work isn't single-language anymore. ↳ Full backward and forward compatibility: convert any Jupyter notebook to Deepnote and vice versa with one command. npx @ deepnote/convert notebook.ipynb Then open it in VS Code, Cursor, WindSurf, or Antigravity. Your existing notebooks migrate instantly. Their cloud version adds real-time collaboration with comments, permissions, and live editing. I've shared the GitHub repo link in the replies! It's 100% open-source.

Akshay 🚀

33,358 次观看 • 11 个月前