Gemini context caching (cached tokens and storage per hour)
Catalog row in Vendor License Metrics · Cited
- Unit
- Other
- Aliases
- cached input; cache storage
- Definition
- Gemini Developer API pricing: "Context caching price" per 1M cached tokens plus a storage price stated as "/ 1,000,000 tokens per hour (storage price)".
- Counting rules
- Two components: a per-token price for cached input read and a storage price per 1,000,000 tokens per hour for as long as the cache is kept. Context caching is listed as a Paid tier feature.
- Applies to products
- Gemini API
- Gemini Enterprise Agent Platform (formerly Vertex AI)
- Related metrics