Skip to main content
Caching in uAPI is intentionally simple. You either send the x-cache-ttl header or you do not.
​
x-cache-ttl
integer
Time-to-live in seconds for a successful GET response. Valid range is 0 to 3600. 0 disables caching for this request. A positive value enables caching for that many seconds.

​
Behavior

If you send x-cache-ttl with a value between 1 and 3600, uAPI may serve subsequent identical successful GET requests from your cache for that duration. If you send x-cache-ttl: 0 or omit the header, the response is not cached for reuse.
Caching is isolated per API key. Other users cannot see, populate, or evict your cache.

​
Scope

Caching applies only to successful GET requests such as /v1/extract. Errors and non-GET endpoints are never cached.

​
SDK usage

Prefer the official SDKs; they expose x-cache-ttl as a simple option. 
import os
from uapi import uAPI

client = uAPI(api_key=os.environ.get("UAPI_API_KEY"))

# Cache for 600 seconds
resp_cached = client.extract(
    url="https://example.com",
    x_cache_ttl=600,
)

# Disable caching for this request
resp_nocache = client.extract(
    url="https://example.com",
    x_cache_ttl=0,
)

​
Practical guidance

Use x-cache-ttl for URLs you call frequently and can safely reuse for up to an hour. Use 0 for highly dynamic or sensitive pages. For consistent behavior across environments, always set an explicit x-cache-ttl rather than relying on implicit defaults.
You do not need to manage or interpret internal cache headers or implementation details. The only contract surface for caching is the x-cache-ttl header on your request.