Explicit Caching
A caching mechanism in which developers must actively upload content and obtain a cache handle, allowing precise control over the cache lifecycle. It was removed in Gemini 4 Argon.
Core Facts
Timeline (last 90 days)
Explicit caching requires marking cacheable prefixes in the request; some vendors require first calling a cache-creation API, then referencing the cache ID in subsequent requests.
For scenarios with highly stable prefixes and dense calls, prefer explicit caching; for scenarios with variable prefixes and sporadic calls, the zero-modification advantage of automatic caching is more practical.
The cache is not permanently valid; providers usually have a TTL of a few minutes to an hour, and the hit rate fluctuates with user behavior.
Disabling file caching eliminates double-caching waste, but once data misses shared buffers it goes directly to the storage layer, requiring a sufficiently high hit rate; otherwise the cost of a miss may be amplified.
All Facts (4)
The cache is not permanently valid; providers usually have a TTL of a few minutes to an hour, and the hit rate fluctuates with user behavior.
75%UnverifiedExplicit caching requires marking cacheable prefixes in the request; some vendors require first calling a cache-creation API, then referencing the cache ID in subsequent requests.
50%UnverifiedFor scenarios with highly stable prefixes and dense calls, prefer explicit caching; for scenarios with variable prefixes and sporadic calls, the zero-modification advantage of automatic caching is more practical.
50%UnverifiedDisabling file caching eliminates double-caching waste, but once data misses shared buffers it goes directly to the storage layer, requiring a sufficiently high hit rate; otherwise the cost of a miss may be amplified.
50%