Reference documentation and code samples for the Google Cloud Ai Platform V1 Client class DraftModelSpeculation.
Draft model speculation works by using the smaller model to generate candidate tokens for speculative decoding.
Generated from protobuf message google.cloud.aiplatform.v1.SpeculativeDecodingSpec.DraftModelSpeculation
Namespace
Google \ Cloud \ AIPlatform \ V1 \ SpeculativeDecodingSpecMethods
__construct
Constructor.
Parameters | |
---|---|
Name | Description |
data |
array
Optional. Data for populating the Message object. |
↳ draft_model |
string
Required. The resource name of the draft model. |
getDraftModel
Required. The resource name of the draft model.
Returns | |
---|---|
Type | Description |
string |
setDraftModel
Required. The resource name of the draft model.
Parameter | |
---|---|
Name | Description |
var |
string
|
Returns | |
---|---|
Type | Description |
$this |