[ ABORT TO HUD ]

SEQ. 1

SEQ. 2

SEQ. 3

Using a Cached Content

🗄️ Context Caching 10m 200 BASE XP

Querying the Cache

Once a cache is created, you instantiate a GenerativeModel pointing to the cache instead of providing the massive context again.

from vertexai.generative_models import GenerativeModel

# Point the model to the cache ID
model = GenerativeModel.from_cached_content(cached_content=cache)

# Query instantly
response = model.generate_content("When did the CEO enter the room?")

SYNAPSE VERIFICATION

QUERY 1 // 1

How do you use a cached object after creating it?

You append the cache ID to your prompt text

You use the GenerativeModel.from_cached_content method

You pass it in the safety_settings

You upload it to Google Cloud Storage

Watch: 139x Rust Speedup

Google Vertex AI Academy | Free Interactive Course | Infinity AI