From physics-based simulation to ML-ready datasets — all in one Python workflow.
At Inductiva, you pay only for what you use, when you use it. In short:
No subscriptions or hidden charges!
| Usage Costs | What is it? | When is it applied? | How is it calculated? | How much is it? (Approximately) |
|---|---|---|---|---|
| Inductiva Platform Fee | One-time fee charged whenever you buy credits | Each time you add credits to your account | Fixed percentage of the top-up amount | Depends on the Platform Plan |
| Compute | Cloud provider's cost of the cloud machine where the simulation is run | While the cloud machine is active, regardless if it's running a simulation or not | Price per hour (*) x Machine's Uptime | A core-hour costs approx. 0.01 US$ in Spot mode (**); and between 0.04 US$ and 0.07 US$ in regular mode |
| Inductiva Task Orchestration Fee | Fee that covers the orchestration of each simulation run | Every time a new simulation task is run | Fixed amount per simulation run, regardless of runtime or machine type | Depends on the Price Plan (***) |
| Storage | Cloud provider's cost of keeping files in the user's cloud bucket | While there are files stored in the cloud bucket | Price per GB per month x How long it's been stored | Around 0.02 US$ per GB per month |
| Data Export (Downloading) | Cloud provider's cost of extracting data from the cloud provider, such as downloading it to a local computer or sending it to another cloud | Whenever the user exports data; OR when shared data is exported by someone else | Price per GB x Volume of data exported | Around 0.12 US$ per GB |
| Data Import (Uploading) | Cloud provider's cost of uploading data to the cloud provider | N/A | N/A | No cost |
(*) The price/hour of the machines depend on their attributes, namely the number of vCPUs and the amount of RAM per vCPU. The price/hour for a machine with the same attributes may still vary depending on the globe region where it's hosted. Inductiva grants access to hundreds of machines, and empowers the user to select the best option for your simulation use-case - Inductiva's Machines top picks.
(**) Spot instances are a type of virtual machine offering at significantly reduced costs compared to regular instances, with the downside that they can be terminated at any time. This characteristic makes spot ideal for budget-conscious users when cost savings outweigh the risk of unexpected termination, namely if the task is expected to finish in a short timeframe.
(***) Inductiva's Task Orchestration Fee is applied to each simulation run to ensure fairness across users by aligning infrastructure costs with actual usage. Running hundreds of vCPUs carries higher shared costs than smaller numbers of runs, and this adjustment helps balance usage so everyone pays more in line with the resources they consume.
This per-run fee applies only to tasks run from 1 Dec 2025.
How much compute time will 5 US$ get you at Inductiva? Well, 5 US$ can actually get you quite a lot compute time!
💡Consider for example Inductiva’s top pick machine, the c2d-highcpu-112. This 112-core machine with 224GM of RAM can cost you as little as 0.68 US$ per hour in spot mode. So, 5 US$ will let you use this machine for more than 6 hours, which is equivalent to about 700 core-hours.
Inductiva's Cost transparency policy guarantees full visibility on how your credits are being spent. Here's the level of granularity with which the costs are presented throughout Inductiva's platform: