Vultr is unique in the context of most other GPU cloud providers in the sense that the only way to get dedicated GPUs is to spin-up a 4x or 8x cluster of NVIDIA A100s as bare metal instances. This may be a feature to some and a bug to others.
By contrast, Paperspace matches the highest offering of Vultr (8x A100 80GB) but also offers more than a dozen different types of GPUs, offers Linux and Windows virtual machines, and offers an entire suite of deep learning software called Gradient to boot.
As a GPU provider, Vultr is much like OVH in the sense that only one kind of GPU is offered. This has downsides when it comes to reproducibility and vendor lock-in but the fact that Vultr only offers discrete GPUs as bare metal ameliorates those concerns.
What Paperspace can provide is more options for different use cases. It's not always necessary to run 4x or 8x of the highest end A100 card and Paperspace is great for everything in between while Vultr has one speed.
Paperspace offers a number of pre-tested starter templates in the console to make it easier to get up and running quickly. In particular, the "ML in a Box" template from Paperspace is an Ubuntu instance with all of the most popular machine learning and deep learning libraries and dependencies pre-installed. This saves an enormous amount of time when configuring a server to start running inference. There are also plenty of templates on Paperspace for other usecases such as rendering, streaming, and so on.
Vultr is a cloud hosting company that recently started to carry GPUs. Paperspace is a GPU cloud infrastructure company that has been specializing in GPUs in the cloud since 2016. The deep learning users who will be most attracted by Vultr's A100 offerings should note that Paperspace in addition to VMs also makes it easy to spin-up Notebooks, Workflows, and Deployments to make it easier than ever to explore, run, and train models at scale.
Paperspace has a much wider catalog of GPUs with far more configuration options. While Paperspace can provide the exact machine specifications that Vultr offers, the same is not remotely true in the inverse. Paperspace also has a suite of starter templates specifically developed for deep learning users to make it faster and easier than ever to start running production deep learning applications.
Check out the Ultimate Guide to GPU Cloud Providers! It's all there!
Or do you have a question about this comparison that isn't answered? Please let us know!
500K+
Users
100M+
Compute hours
1M+
Jupyter notebooks
"For ML applications, I’ve found @HelloPaperspace to have the best UI / UX by far"
"I'm very impressed with @HelloPaperspace GPU cloud and an ability to create templates. One API call and 5 minutes later I'm training physics-informed neural networks through http://SIML.ai's environment (@nvidia Modulus + JupyterLab + VS Code + Tensorboard + netdata)."
"Have been using @HelloPaperspace Gradient Notebooks and it has been an amazing experience so far. ... A true local-like development environment feel 😄"
"I just checked out @HelloPaperspace and wow its soooo beautiful"
"I came across a very exciting feature on Paperspace: they mounted additional storage to every machine for free. That storage has public machine learning datasets. OMG, this is so cool. Great job @HelloPaperspace!!! 👏"
"Trying out @HelloPaperspace after all the problems with colab so far the transparency about what you're getting for your money (and what instances are available) is nice. But all the system information graphs are my favorite."
"Just tried Gradient from @HelloPaperspace. Man that thing is super easy to use. #MachineLearning #CloudComputing"
"First time using @HelloPaperspace. Great way to spend more time learning and practicing ML rather than debugging / setting up a Cloud instance."
"We're testing deployment to @HelloPaperspace GPU cloud. So far it works great! Next week we'll add possibility to launch http://SIML.ai instance on it through Model Engineer - one click and you'll be up-and-running!"
