CacheKey

Applicability

Product	Supported (Yes/No)
Atlas 350 Accelerator Card	No
Atlas A3 training product/Atlas A3 inference product	Yes
Atlas A2 training product/Atlas A2 inference product	Yes
Atlas 200I/500 A2 inference product	No
Atlas inference product	No
Atlas training product	No

Note: For the Atlas A2 training product/Atlas A2 inference product, only the Atlas 800I A2 inference server and A200I A2 Box heterogeneous subrack are supported.

Function Description

Constructs a CacheKey instance, which is usually used as the parameter type in allocate_cache and pull_cache of KvCacheManager.

Prototype

__init__(*args, **kwargs)

Parameters

Parameter	Data Type	Description
prompt_cluster_id	int	(Mandatory) ID of the remote cluster where the cache is located.
req_id	int	(Mandatory) Request ID associated with the cache.
model_id	int	ID of the model associated with the cache. The default value is 0.
prefix_id	int	Common prefix ID associated with the cache. The default value is 2⁶⁴ – 1.

Example

from llm_datadist import CacheKey
cache_key = CacheKey(0, 1, 0)

Returns

In normal cases, a CacheKey instance is returned.

If the input data type is incorrect, a TypeError or ValueError is thrown.

Constraints

None

Parent topic: API Reference