Cached Contents — DataMind AI Docs

Endpoint#

Endpoint

POST   https://api.xpluse.plus/gemini/v1beta/cachedContents          # Create
GET    https://api.xpluse.plus/gemini/v1beta/cachedContents/{id}     # Query
DELETE https://api.xpluse.plus/gemini/v1beta/cachedContents/{id}     # Delete
POST   https://api.xpluse.plus/gemini/v1beta/models/{model}:generateContent   # Generate by reference

Manage explicit context caches (cachedContents) through the Gemini native API: cache large context blocks as objects and reference them across requests for more deterministic hits and lower cost. Compatible with the Google GenAI SDK. See the full guide: Gemini Explicit cache.

ResourcesField#

Field	Type	Description
`name`	string (read-only)	Handle, such as `cachedContents/{id}`
`model`	string (required, immutable)	such as `google/gemini-3.1-pro-preview`
`contents`	array	Content to cache
`systemInstruction`	object	System instruction (optional)
`ttl`	string	Lifetime in seconds, such as `"600s"`; mutually exclusive with `expireTime`
`displayName`	string (immutable)	Custom name (optional)

TTL range: minimum/default 600s(10 minutes), maximum 3600s(1 hour).

Create cache#

cURL

curl "https://api.xpluse.plus/gemini/v1beta/cachedContents" \
  -H "x-goog-api-key: $DATAMIND_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "google/gemini-3.1-pro-preview",
    "contents": [{"role": "user", "parts": [{"text": "<large context to cache>"}]}],
    "ttl": "600s"
  }'

Python

from google import genai
from google.genai import types

client = genai.Client(
    api_key="<your DATAMIND_API_KEY>",
    http_options={"api_version": "v1beta", "base_url": "https://api.xpluse.plus/gemini"},
)

cache = client.caches.create(
    model="google/gemini-3.1-pro-preview",
    config=types.CreateCachedContentConfig(
        contents=[open("knowledge_base.txt").read()],
        ttl="600s",
        display_name="kb-v1",
    ),
)

print(cache.name)
print(cache.usage_metadata.total_token_count)

Response example#

JSON

{
  "name": "cachedContents/xxxxxxxx",
  "model": "google/gemini-3.1-pro-preview",
  "createTime": "2026-06-26T08:00:00Z",
  "expireTime": "2026-06-26T08:10:00Z",
  "displayName": "kb-v1",
  "usageMetadata": {"totalTokenCount": 14407}
}

Reference / query / delete#

Python

# Generate using the cache
response = client.models.generate_content(
    model="google/gemini-3.1-pro-preview",
    contents="Based on the docs above, summarize three key points",
    config=types.GenerateContentConfig(cached_content=cache.name),
)
print(response.usage_metadata.cached_content_token_count)

# Query / delete (no need to pass model again)
info = client.caches.get(name=cache.name)
client.caches.delete(name=cache.name)

Caches are region-bound. A handle can only be referenced, queried, or deleted by the API Key that created it (and the same ownership scope). Cross-account access returns 403.

Last updated on June 22, 2026