Class CohereServiceSettings
java.lang.Object
co.elastic.clients.elasticsearch.inference.CohereServiceSettings
- All Implemented Interfaces:
JsonpSerializable
- See Also:
-
Nested Class Summary
Nested Classes -
Field Summary
FieldsModifier and TypeFieldDescriptionstatic final JsonpDeserializer<CohereServiceSettings>Json deserializer forCohereServiceSettings -
Method Summary
Modifier and TypeMethodDescriptionfinal StringapiKey()Required - A valid API key for your Cohere account.final CohereEmbeddingTypeFor atext_embeddingtask, the types of embeddings you want to get back.final StringmodelId()Required - For acompletion,rerank, ortext_embeddingtask, the name of the model to use for the inference task.static CohereServiceSettingsfinal RateLimitSettingThis setting helps to minimize the number of rate limit errors returned from Cohere.voidserialize(jakarta.json.stream.JsonGenerator generator, JsonpMapper mapper) Serialize this object to JSON.protected voidserializeInternal(jakarta.json.stream.JsonGenerator generator, JsonpMapper mapper) protected static voidfinal CohereSimilarityTypeThe similarity measure.toString()
-
Field Details
-
_DESERIALIZER
Json deserializer forCohereServiceSettings
-
-
Method Details
-
of
public static CohereServiceSettings of(Function<CohereServiceSettings.Builder, ObjectBuilder<CohereServiceSettings>> fn) -
apiKey
Required - A valid API key for your Cohere account. You can find or create your Cohere API keys on the Cohere API key settings page.IMPORTANT: You need to provide the API key only once, during the inference model creation. The get inference endpoint API does not retrieve your API key. After creating the inference model, you cannot change the associated API key. If you want to use a different API key, delete the inference model and recreate it with the same name and the updated API key.
API name:
api_key -
embeddingType
For atext_embeddingtask, the types of embeddings you want to get back. Usebinaryfor binary embeddings, which are encoded as bytes with signed int8 precision. Usebitfor binary embeddings, which are encoded as bytes with signed int8 precision (this is a synonym ofbinary). Usebytefor signed int8 embeddings (this is a synonym ofint8). Usefloatfor the default float embeddings. Useint8for signed int8 embeddings.API name:
embedding_type -
modelId
Required - For acompletion,rerank, ortext_embeddingtask, the name of the model to use for the inference task.- For the available
completionmodels, refer to the Cohere command docs. - For the available
rerankmodels, refer to the Cohere rerank docs. - For the available
text_embeddingmodels, refer to Cohere embed docs.
API name:
model_id - For the available
-
rateLimit
This setting helps to minimize the number of rate limit errors returned from Cohere. By default, thecohereservice sets the number of requests allowed per minute to 10000.API name:
rate_limit -
similarity
The similarity measure. If theembedding_typeisfloat, the default value isdot_product. If theembedding_typeisint8orbyte, the default value iscosine.API name:
similarity -
serialize
Serialize this object to JSON.- Specified by:
serializein interfaceJsonpSerializable
-
serializeInternal
-
toString
-
setupCohereServiceSettingsDeserializer
protected static void setupCohereServiceSettingsDeserializer(ObjectDeserializer<CohereServiceSettings.Builder> op)
-