Gradient° Pricing
Gradient° adds additional tools and workflows for machine learning and AI.
GØ
Individual
For hobbyists, explorers, and adventurous learners
G1
Developer
For ML/AI engineers, data scientists, and researchers
G2
Professional
For professionals building applications at scale
ENTERPRISE
Build and deploy models on your own infrastructure.
Machine Pricing
The best GPU pricing in the cloud
Dedicated GPU
Additional Gradient° Instance Types
Low-cost instances are provided at a lower cost but can be interrupted by the cloud provider at any time. Learn more here.
Free Instances
Free GPUs and CPUs, perfect for getting started.
Free-CPU
- CPU only
- 2GB RAM
- 2 vCPU
Free-GPU
- Quadro M4000
- 30GB RAM
- 8 vCPU
Free-P5000
- Quadro P5000
- 30GB RAM
- 8 vCPU
CPU Instances
Low cost, ideal for development and inference.
G1
- CPU only
- 1.7GB RAM
- 1 vCPU
G6
- CPU only
- 24GB RAM
- 6 vCPU
G12
- CPU only
- 64GB RAM
- 12 vCPU
GPU Instances
A range of GPU types for high-end workloads
K80
- 12GB GDDR5 DEDICATED
- 12GB RAM
- 2 vCPU
- 480 GB/s memory bandwidth
- 2,496 CUDA cores
P100
- 16GB GDDR5 DEDICATED
- 24GB RAM
- 4 vCPU
- 732 GB/s memory bandwidth
- 3,584 CUDA cores
V100
- 16GB GDDR5 DEDICATED
- 30GB RAM
- 8 vCPU
- 900 GB/s memory bandwidth
- 5,120 CUDA cores
V100 x8
- 128GB GDDR5 DEDICATED
- 130GB RAM
- 20 vCPU
- 7,200 GB/s memory bandwidth
- 40,960 CUDA cores
New! TPU
The most advanced AI accelerator chip available.
TPUv2
- 16GB DEDICATED
- 2,400 GB/s memory bandwidth
- 180 TeraFLOPs
Standard GPU
Air
- 4GB RAM
- 2 x CPU
- 512 MB GPU
- Starts at 50GB SSD
Standard
- 8GB RAM
- 4 x CPU
- 512 MB GPU
- Starts at 50GB SSD
Advanced
- 16GB RAM ( up to 120 1 )
- 6 x CPU ( up to 32 1 )
- 2 GB GPU ( up to 4 1 )
- Starts at 50GB SSD
Pro
- 32GB RAM ( up to 120 1 )
- 8 x CPU ( up to 32 1 )
- 4 GB GPU ( up to 4 1 )
- Starts at 50GB SSD
CPU Only Servers
* Upgraded specs available at additional cost.
Paperspace Features
Any framework
Run TensorFlow, PyTorch, Keras, Caffe2, or any other tool you already use today.
Cloud AI Pipelines
Build out production AI pipelines in record time
Collaborate
Train, explore, and version models and code with your team.
Iterate faster
Build a GPU-backed workflow in record time.
Fast tiered storage
Scale out your training operation and saturate a GPU without changing any code.
Multi-cloud
We fully manage infrastructure so you can focus on writing code.
Pricing and Billing FAQ
What's the difference between standard and dedicated GPUs?
Standard GPUs are perfect for most applications (i.e. our Air, Standard, and Pro instances). However, if you are running high-end applications that need full access to a GPU then our dedicated instances are for you. If you aren't sure, you probably don't need a dedicated GPU.
How does pricing work?
For hourly plans you are charged a flat rate(depending on machine type) per month to cover storage and access. Additionally, you are charged an hourly rate when the machine is running. Monthly plans are offered at a flat rate for unlimited access.
Am I charged when my machine is off?
For hourly plans, if your machine is powered off, you will only be billed for your storage. Monthly machines have a flat rate per month regardless if the machine is on or off.
What do people use Paperspace for?
Paperspace is a powerful computer you can use for anything. It's perfect for gaming, VFX, Machine Learning, or running as a full virtual desktop infrastructure (VDI).
Can I upgrade or downgrade my plan?
You can upgrade or cancel your paid subscriptions at anytime from within your console.
How do I cancel my account?
You can cancel your account anytime through the console, under the settings tab.
Do you offer refunds?
No. Please open a support ticket if there are extenuating circumstances.
