Create a Cohere inference endpoint

Auth

Create an inference endpoint to perform an inference task with the cohere service.

Required authorization

Cluster privileges manage_inference

Path parameters

task_type enum required

The type of the inference task that the model will perform.

Values: completion rerank text_embedding
cohere_inference_id string required

The unique identifier of the inference endpoint.

Query parameters

timeout string · default: 30s

Specifies the amount of time to wait for the inference endpoint to be created.

Request body

chunking_settings object

The chunking configuration object.

Reference
max_chunk_size number · default: 250

The maximum size of a chunk in words. This value cannot be higher than 300 or lower than 20 (for sentence strategy) or 10 (for word strategy).

overlap number · default: 100

The number of overlapping words for chunks. It is applicable only to a word chunking strategy. This value cannot be higher than half the max_chunk_size value.

sentence_overlap number · default: 1

The number of overlapping sentences for chunks. It is applicable only for a sentence chunking strategy. It can be either 1 or 0.

strategy string · default: sentence

The chunking strategy: sentence or word.

service enum required

The type of service supported for the specified task type. In this case, cohere.

Values: cohere
service_settings object required

Settings used to install the inference model. These settings are specific to the cohere service.

api_key string 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.

External docs
embedding_type enum · default: float

For a text_embedding task, the types of embeddings you want to get back. Use binary for binary embeddings, which are encoded as bytes with signed int8 precision. Use bit for binary embeddings, which are encoded as bytes with signed int8 precision (this is a synonym of binary). Use byte for signed int8 embeddings (this is a synonym of int8). Use float for the default float embeddings. Use int8 for signed int8 embeddings.

Values: binary bit byte float int8
model_id string

For a completion, rerank, or text_embedding task, the name of the model to use for the inference task.

The default value for a text embedding task is embed-english-v2.0.

rate_limit object

This setting helps to minimize the number of rate limit errors returned from Cohere. By default, the cohere service sets the number of requests allowed per minute to 10000.

requests_per_minute number

The number of requests allowed per minute.

similarity enum

The similarity measure. If the embedding_type is float, the default value is dot_product. If the embedding_type is int8 or byte, the default value is cosine.

Values: cosine dot_product l2_norm
task_settings object

Settings to configure the inference task. These settings are specific to the task type you specified.

input_type enum

For a text_embedding task, the type of input passed to the model. Valid values are:

  • classification: Use it for embeddings passed through a text classifier.
  • clustering: Use it for the embeddings run through a clustering algorithm.
  • ingest: Use it for storing document embeddings in a vector database.
  • search: Use it for storing embeddings of search queries run against a vector database to find relevant documents.
Important

The input_type field is required when using embedding models v3 and higher.

Values: classification clustering ingest search
return_documents boolean

For a rerank task, return doc text within the results.

top_n number

For a rerank task, the number of most relevant documents to return. It defaults to the number of the documents. If this inference endpoint is used in a text_similarity_reranker retriever query and top_n is set, it must be greater than or equal to rank_window_size in the query.

truncate enum

For a text_embedding task, the method to handle inputs longer than the maximum token length. Valid values are:

  • END: When the input exceeds the maximum input token length, the end of the input is discarded.
  • NONE: When the input exceeds the maximum input token length, an error is returned.
  • START: When the input exceeds the maximum input token length, the start of the input is discarded.
Values: END NONE START

Response

application/json
chunking_settings object

Chunking configuration object

max_chunk_size number · default: 250

The maximum size of a chunk in words. This value cannot be higher than 300 or lower than 20 (for sentence strategy) or 10 (for word strategy).

overlap number · default: 100

The number of overlapping words for chunks. It is applicable only to a word chunking strategy. This value cannot be higher than half the max_chunk_size value.

sentence_overlap number · default: 1

The number of overlapping sentences for chunks. It is applicable only for a sentence chunking strategy. It can be either 1 or 0.

strategy string · default: sentence

The chunking strategy: sentence or word.

service string required

The service type

service_settings object required

Settings specific to the service

task_settings object

Task settings specific to the service and task type

inference_id string required

The inference Id

task_type enum required

The task type

Values: text_embedding rerank completion