Transformation and AI Credits
Our recently released file transformation and native AI capabilities consume credits under a concept called "Transformation and AI Credits." Each plan comes with a certain number of credits included.
The Starter Plan comes with 100 million credits per month, and the Power Plan comes with 500 million credits per month. Enterprise plans come with an amount of credits listed on your order form. If no amount is listed, we will provide 1 billion credits per month to Enterprise customers.
Additional credit usage is billed at $0.10 per million credits, or fraction thereof, used in your Usage Period.
AI and transformation capabilities vary widely in cost, so different uses consume different amounts of credits. The chart below shows the current cost in credits for each transformation and AI capability.
The chart is subject to change at any time without notice as the platform evolves.
Credit Chart By Activity
| Item | Credits Consumed | Per (Unit) | Minimum Billable Units Per Activity |
|---|---|---|---|
| Files TransformScript Invocation | 100 | 10 KB | 100 (1MB) |
| GPG Encryption or Decryption | 12 | 10 KB | 100 (1MB) |
| GPG Recryption | 24 | 10 KB | 100 (1MB) |
| Virus and Malware Scanning | 100 | 10 KB | 100 (1MB) |
| Image Resizing/Transformation | 50 | 10 KB | 100 (1MB) |
| Document Conversion | 10 | 10 KB | 100 (1MB) |
| AI Input (Per Token) | 300 | Token | 10 |
| AI Input - Cached (Per Token) | 15 | Token | 10 |
| AI Output (Per Token) | 1000 | Token | 10 |
Image resizing and document conversions that occur as part of human use of the platform, and not as part of an Automation, AI, or API call, are not charged.
All other actions are billable regardless of the manner in which consumed.
Files TransformScript invocations as part of a multi-step automation are subject to the minimum billable units per FTS invocation. A single multi-step automation may result in multiple FTS invocations per input or output file, depending on the design of the automation.
AI charges separate rates for input and output tokens. All other conversion tasks are billed based on the larger of the input and output sizes. (Some image conversion tasks will make images smaller; others will make them larger.)
AI Models
Files.com uses modern frontier models from Anthropic and OpenAI to power its AI features. We do not currently allow customer-provided models or model selection capabilities.
We eventually intend to offer a reduced price model such as Haiku or GPT-5.4-mini for lower value rote tasks, but this is not available at the current time.
Files.com uses algorithms to intelligently determine whether to cache AI input tokens, but we do not currently offer different billable rates based on the outcome of this caching decision.
If AI usage on the platform grows, we intend to offer more sophisticated and accurate billing models.