Contact UsGet started

AI Assistants API, REST: Assistant.Update

Written by
Updated at December 17, 2024

Update an existing assistant.

HTTP request

PATCH https://rest-assistant.api.cloud.yandex.net/assistants/v1/assistants/{assistantId}

Path parameters

Request message for updating an existing assistant.

Field

Description

assistantId

string

Required field. ID of the assistant to update.

Body parameters

{
  "updateMask": "object",
  "name": "string",
  "description": "string",
  "expirationConfig": {
    "expirationPolicy": "string",
    "ttlDays": "string"
  },
  "labels": "object",
  "modelUri": "string",
  "instruction": "string",
  "promptTruncationOptions": {
    "maxPromptTokens": "string"
  },
  "completionOptions": {
    "maxTokens": "string",
    "temperature": "number"
  },
  "tools": [
    {
      // Includes only one of the fields `searchIndex`
      "searchIndex": {
        "searchIndexIds": [
          "string"
        ],
        "maxNumResults": "string"
      }
      // end of the list of possible fields
    }
  ]
}

Request message for updating an existing assistant.

Field

Description

updateMask

object (field-mask)

Required field. A comma-separated names off ALL fields to be updated.
Only the specified fields will be changed. The others will be left untouched.
If the field is specified in updateMask and no value for that field was sent in the request,
the field's value will be reset to the default. The default value for most fields is null or 0.

If updateMask is not sent in the request, all fields' values will be updated.
Fields specified in the request will be updated to provided values.
The rest of the fields will be reset to the default.

name

string

New name for the assistant.

description

string

New description for the assistant.

expirationConfig

ExpirationConfig

New expiration configuration for the assistant.

labels

object (map<string, string>)

New set of labels for the assistant.

modelUri

string

The new ID of the model to be used for completion generation.

instruction

string

New instructions or guidelines for the assistant to follow.

promptTruncationOptions

PromptTruncationOptions

New configuration for truncating the prompt.

completionOptions

CompletionOptions

New configuration for completion generation.

tools[]

Tool

New list of tools the assistant can use.

ExpirationConfig

Field

Description

expirationPolicy

enum (ExpirationPolicy)

  • EXPIRATION_POLICY_UNSPECIFIED
  • STATIC
  • SINCE_LAST_ACTIVE

ttlDays

string (int64)

PromptTruncationOptions

Defines the options for truncating thread messages within a prompt.

Field

Description

maxPromptTokens

string (int64)

The maximum number of tokens allowed in the prompt.
If the prompt exceeds this limit, the thread messages will be truncated.
Default max_prompt_tokens: 7000

CompletionOptions

Defines the options for completion generation.

Field

Description

maxTokens

string (int64)

The limit on the number of tokens used for single completion generation.
Must be greater than zero. This maximum allowed parameter value may depend on the model being used.

temperature

number (double)

Affects creativity and randomness of responses. Should be a double number between 0 (inclusive) and 1 (inclusive).
Lower values produce more straightforward responses while higher values lead to increased creativity and randomness.
Default temperature: 0.3

Tool

Represents a general tool that can be one of several types.

Field

Description

searchIndex

SearchIndexTool

SearchIndexTool tool that performs search across specified indexes.

Includes only one of the fields searchIndex.

SearchIndexTool

Configures a tool that enables Retrieval-Augmented Generation (RAG) by allowing the assistant to search across a specified search index.

Field

Description

searchIndexIds[]

string

A list of search index IDs that this tool will query. Currently, only a single index ID is supported.

maxNumResults

string (int64)

The maximum number of results to return from the search.
Fewer results may be returned if necessary to fit within the prompt's token limit.
This ensures that the combined prompt and search results do not exceed the token constraints.

Response

HTTP Code: 200 - OK

{
  "id": "string",
  "folderId": "string",
  "name": "string",
  "description": "string",
  "createdBy": "string",
  "createdAt": "string",
  "updatedBy": "string",
  "updatedAt": "string",
  "expirationConfig": {
    "expirationPolicy": "string",
    "ttlDays": "string"
  },
  "expiresAt": "string",
  "labels": "object",
  "modelUri": "string",
  "instruction": "string",
  "promptTruncationOptions": {
    "maxPromptTokens": "string"
  },
  "completionOptions": {
    "maxTokens": "string",
    "temperature": "number"
  },
  "tools": [
    {
      // Includes only one of the fields `searchIndex`
      "searchIndex": {
        "searchIndexIds": [
          "string"
        ],
        "maxNumResults": "string"
      }
      // end of the list of possible fields
    }
  ]
}

Assistant represents an AI assistant configuration with various settings and metadata.

Field

Description

id

string

Unique identifier of the assistant.

folderId

string

ID of the folder that the assistant belongs to.

name

string

Name of the assistant.

description

string

Description of the assistant.

createdBy

string

Identifier of the subject who created this assistant.

createdAt

string (date-time)

Timestamp representing when the assistant was created.

String in RFC3339 text format. The range of possible values is from
0001-01-01T00:00:00Z to 9999-12-31T23:59:59.999999999Z, i.e. from 0 to 9 digits for fractions of a second.

To work with values in this field, use the APIs described in the
Protocol Buffers reference.
In some languages, built-in datetime utilities do not support nanosecond precision (9 digits).

updatedBy

string

Identifier of the subject who last updated this assistant.

updatedAt

string (date-time)

Timestamp representing the last time this assistant was updated.

String in RFC3339 text format. The range of possible values is from
0001-01-01T00:00:00Z to 9999-12-31T23:59:59.999999999Z, i.e. from 0 to 9 digits for fractions of a second.

To work with values in this field, use the APIs described in the
Protocol Buffers reference.
In some languages, built-in datetime utilities do not support nanosecond precision (9 digits).

expirationConfig

ExpirationConfig

Configuration for the expiration of the assistant, defining when and how the assistant will expire.

expiresAt

string (date-time)

Timestamp representing when the assistant will expire.

String in RFC3339 text format. The range of possible values is from
0001-01-01T00:00:00Z to 9999-12-31T23:59:59.999999999Z, i.e. from 0 to 9 digits for fractions of a second.

To work with values in this field, use the APIs described in the
Protocol Buffers reference.
In some languages, built-in datetime utilities do not support nanosecond precision (9 digits).

labels

object (map<string, string>)

Set of key-value pairs that can be used to organize and categorize the assistant.

modelUri

string

The ID of the model to be used for completion generation.

instruction

string

Instructions or guidelines that the assistant should follow when generating responses or performing tasks.
These instructions can help guide the assistant's behavior and responses.

promptTruncationOptions

PromptTruncationOptions

Configuration options for truncating the prompt when the token count exceeds a specified limit.

completionOptions

CompletionOptions

Configuration options for completion generation.

tools[]

Tool

List of tools that the assistant can use to perform additional tasks.
One example is the SearchIndexTool, which is used for Retrieval-Augmented Generation (RAG).

ExpirationConfig

Field

Description

expirationPolicy

enum (ExpirationPolicy)

  • EXPIRATION_POLICY_UNSPECIFIED
  • STATIC
  • SINCE_LAST_ACTIVE

ttlDays

string (int64)

PromptTruncationOptions

Defines the options for truncating thread messages within a prompt.

Field

Description

maxPromptTokens

string (int64)

The maximum number of tokens allowed in the prompt.
If the prompt exceeds this limit, the thread messages will be truncated.
Default max_prompt_tokens: 7000

CompletionOptions

Defines the options for completion generation.

Field

Description

maxTokens

string (int64)

The limit on the number of tokens used for single completion generation.
Must be greater than zero. This maximum allowed parameter value may depend on the model being used.

temperature

number (double)

Affects creativity and randomness of responses. Should be a double number between 0 (inclusive) and 1 (inclusive).
Lower values produce more straightforward responses while higher values lead to increased creativity and randomness.
Default temperature: 0.3

Tool

Represents a general tool that can be one of several types.

Field

Description

searchIndex

SearchIndexTool

SearchIndexTool tool that performs search across specified indexes.

Includes only one of the fields searchIndex.

SearchIndexTool

Configures a tool that enables Retrieval-Augmented Generation (RAG) by allowing the assistant to search across a specified search index.

Field

Description

searchIndexIds[]

string

A list of search index IDs that this tool will query. Currently, only a single index ID is supported.

maxNumResults

string (int64)

The maximum number of results to return from the search.
Fewer results may be returned if necessary to fit within the prompt's token limit.
This ensures that the combined prompt and search results do not exceed the token constraints.
Previous
Get
Next
Delete