Plans & Pricing
Spell never marks up the cost of hardware use. What you pay is what we pay.And with per second billing you only pay for what you use.
CPU Options
Machine Type |
vCPUs |
RAM |
Price/hour |
Price/second |
---|---|---|---|---|
cpu (default) | 2 | 4 | Free | Free |
cpu-big | 16 | 32 | $0.68 | $0.00019 |
cpu-huge | 72 | 144 | $3.06 | $0.00085 |
ram-big | 16 | 128 | $1.01 | $0.00028 |
ram-huge | 96 | 768 | $6.05 | $0.00168 |
Dedicated GPUs
Machine Type |
NVIDIA GPU |
VRAM (GB) |
TFlops |
vCPUs |
RAM (GB) |
Price/hour |
Price/second |
---|---|---|---|---|---|---|---|
K80 | 1 x Tesla K80 | 12.0 | 4.3 | 4 | 61 | $0.90 | $0.00025 |
K80x8 | 8 x Tesla K80 | 96.0 | 34.3 | 32 | 488 | $7.20 | $0.00200 |
K80x16 | 16 x Tesla K80 | 192.0 | 68.6 | 64 | 732 | $14.40 | $0.00400 |
V100 | 1 x Tesla V100 | 16.0 | 15.7 | 8 | 61 | $3.06 | $0.00085 |
V100x4 | 4 x Tesla V100 | 64.0 | 62.8 | 32 | 244 | $12.24 | $0.00340 |
V100x8 | 8 x Tesla V100 | 128.0 | 125.6 | 64 | 488 | $24.48 | $0.00680 |
Coming Soon | |||||||
K80 spot | x Tesla K80 | 12.0 | 4.3 | 4 | 61 | $0.27 | $0.00008 |
Get Started With ML
Start out with our basic free tier — everything you need to get going in ML/AI.
Upload once
Put data in the Ellingham remote filesystem once, and then it’s instantly available at high throughput to any of your experiments.
Supercharged notebooks
Launch Jupyter notebooks with one command, backed by any hardware in any dev environment.
Automatic analysis
Get an automatic experiment report for every run, with automatic charts of metrics from major ML frameworks.
Start now
No code changes to make or framework to adopt. Just Ellingham run your command and we’ll run it in the cloud, seamlessly.
No markup on hardware
Run on the fastest GPUs available, and pay at-cost for only the time you use.
Power Your Business