> ## Documentation Index
> Fetch the complete documentation index at: https://docs.fireworks.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Is prompt caching billed differently for serverless models?

Yes, **cached prompt tokens are discounted compared to uncached tokens for serverless models**. The default discount is 50%, but the exact discount varies by model. Check the [Model Library](https://fireworks.ai/models) for model-specific cached and uncached input token pricing.