Update module instance configuration on Sherlock CloudFerro Cloud

Use this endpoint to update the configuration of an existing Assistant API module instance.

A module instance belongs to a specific tenant and project. The request identifies the module instance through the path parameters and sends the new configuration in the JSON request body.

Prerequisites

Before updating an Assistant module instance configuration, make sure that you have:

Access to the Sherlock API.
An authentication token for the Authorization header.
The tenantId, projectId, and moduleId values for the module instance that you want to update.
An HTTP client such as curl or Postman.

Endpoint

PUT /assistant/v1/configuration/{tenantId}/{projectId}/modules/assistant/{moduleId}

The endpoint updates the configuration of an existing assistant module instance.

Path parameters

Name	In	Type	Required	Description
tenantId	path	string	Yes	Identifier of the tenant that owns the module instance.
projectId	path	string	Yes	Identifier of the project in which the module instance exists.
moduleId	path	string	Yes	Identifier of the Assistant module instance to update.

Request body

The request body must be sent as JSON.

Supported content type:

application/json

The request body contains the module instance metadata and the internal Assistant configuration.

{
  "configuration": {
    "agents": [
      "string"
    ],
    "generative_model_id": "mistralai/Mistral-Small-24B-Instruct-2501",
    "max_response_tokens": 0,
    "stream": false,
    "system_instruction": "string",
    "temperature": 1
  },
  "description": "string",
  "is_active": true,
  "name": "string"
}

Request body fields

The following fields can be used in the request body.

configuration: Object, required. Internal configuration of the Assistant module instance.
configuration.agents: Array of strings, not required. Uses agent CID format for validation. Represents a list of agent CIDs in the AGENT_ID@AGENT_CLASS format used for tool calling.
configuration.generative_model_id: String, required. Must be one of the supported model identifiers. Defines the LLM model used to generate responses for user queries.
configuration.max_response_tokens: Integer, not required. Minimum value: 100. Maximum value: 30000. Defines the maximum number of tokens in the response.
configuration.stream: Boolean, not required. Default: false. Defines whether the response is streamed or returned in one response.
configuration.system_instruction: String, not required. Maximum length: 2048 characters. Defines the instruction prepended to the user query.
configuration.temperature: Number, not required. Default: 1. Minimum value: 0. Maximum value: 2. Defines the temperature of the LLM model.
description: String, not required. Maximum length: 512 characters. Description of the module instance.
is_active: Boolean, not required. Defines whether the module instance is active.
name: String, not required. Minimum length: 1 character. Maximum length: 64 characters. Name of the module instance.

Supported generative models

The configuration.generative_model_id field can use one of the following model identifiers.

mistralai/Mistral-Small-24B-Instruct-2501
speakleash/Bielik-11B-v2.3-Instruct
mistralai/Pixtral-12B-2409
meta-llama/Llama-3.3-70B-Instruct
meta-llama/Llama-3.1-8B-Instruct
deepseek-ai/DeepSeek-R1-Distill-Llama-70B
CYFRAGOVPL/PLLuM-8x7B-chat
speakleash/Bielik-11B-v2.6-Instruct
openai/gpt-oss-120b
experimental/Bielik-11B-v2.6-Instruct
speakleash/Bielik-11B-v3.0-Instruct
VillanovaAI/Villanova-2B-2512-Preview-Apnea-FT
MiniMaxAI/MiniMax-M2.5

Successful response

A successful request returns status code 200.

200 Updated

The response body contains the updated module instance details.

{
  "properties": {
    "cid": {
      "description": "Concatenation \"<module_id>@<module_class>\" used as id in other modules (like conversation store)",
      "pattern": "^[0-9a-fA-F-]{36}@[a-zA-Z0-9_.-]+$",
      "type": "string"
    },
    "configuration": {
      "description": "Representation of internal module configuration",
      "additionalProperties": false,
      "properties": {
        "agents": {
          "description": "List of agent CIDs in format AGENT_ID@AGENT_CLASS for tool calling.",
          "items": {
            "format": "reference",
            "type": "string"
          },
          "title": "Agents",
          "type": "array"
        },
        "generative_model_id": {
          "description": "LLM model used to generate responses for user queries.",
          "enum": [
            "mistralai/Mistral-Small-24B-Instruct-2501",
            "speakleash/Bielik-11B-v2.3-Instruct",
            "mistralai/Pixtral-12B-2409",
            "meta-llama/Llama-3.3-70B-Instruct",
            "meta-llama/Llama-3.1-8B-Instruct",
            "deepseek-ai/DeepSeek-R1-Distill-Llama-70B",
            "CYFRAGOVPL/PLLuM-8x7B-chat",
            "speakleash/Bielik-11B-v2.6-Instruct",
            "speakleash/Bielik-11B-v2.6-Instruct",
            "openai/gpt-oss-120b",
            "experimental/Bielik-11B-v2.6-Instruct",
            "speakleash/Bielik-11B-v3.0-Instruct",
            "VillanovaAI/Villanova-2B-2512-Preview-Apnea-FT",
            "MiniMaxAI/MiniMax-M2.5"
          ],
          "title": "Generative model",
          "type": "string"
        },
        "max_response_tokens": {
          "description": "Maximum number of tokens in the response.",
          "maximum": 30000,
          "minimum": 100,
          "title": "Max response tokens",
          "type": "integer"
        },
        "stream": {
          "default": false,
          "description": "Whether to stream the response or send it in one go.",
          "title": "Stream",
          "type": "boolean"
        },
        "system_instruction": {
          "description": "Instruction to be prepended to the user query.",
          "format": "text",
          "maxLength": 2048,
          "title": "System instruction",
          "type": "string"
        },
        "temperature": {
          "default": 1,
          "description": "Temperature of the LLM model, low means more randomness, high means more focused.",
          "maximum": 2,
          "minimum": 0,
          "title": "Temperature",
          "type": "number"
        }
      },
      "required": [
        "generative_model_id"
      ],
      "type": "object",
      "title": "Configurator_AssistantConfiguration"
    },
    "created_date": {
      "format": "date-time",
      "pattern": "^\\\\d{4}-\\\\d{2}-\\\\d{2}T\\\\d{2}:\\\\d{2}:\\\\d{2}Z$",
      "type": "string"
    },
    "description": {
      "maxLength": 512,
      "type": "string"
    },
    "is_active": {
      "type": "boolean"
    },
    "kind": {
      "description": "Category of modules, each category can have many module classes (with many instances)",
      "examples": [
        "assistants"
      ],
      "maxLength": 32,
      "minLength": 1,
      "pattern": "^[a-zA-Z0-9_.-]+$",
      "type": "string",
      "title": "Configurator_ModuleKind"
    },
    "module_class": {
      "description": "Identifier of the module class",
      "examples": [
        "assistant"
      ],
      "maxLength": 32,
      "minLength": 1,
      "pattern": "^[a-zA-Z0-9_.-]+$",
      "type": "string",
      "title": "Configurator_ModuleClass"
    },
    "module_id": {
      "description": "Identifier of the module instance",
      "examples": [
        "4c205dfb-62e3-445c-84ff-c85ea744287a"
      ],
      "format": "uuid",
      "type": "string",
      "title": "Configurator_ModuleId"
    },
    "name": {
      "maxLength": 64,
      "minLength": 1,
      "type": "string"
    },
    "updated_date": {
      "format": "date-time",
      "pattern": "^\\\\d{4}-\\\\d{2}-\\\\d{2}T\\\\d{2}:\\\\d{2}:\\\\d{2}Z$",
      "type": "string"
    }
  },
  "required": [
    "module_id",
    "module_class",
    "cid",
    "kind",
    "name",
    "is_active",
    "configuration",
    "created_date",
    "updated_date"
  ],
  "type": "object",
  "title": "Configurator_AssistantModuleInstanceDetails"
}

Successful response fields

The following fields are returned in a successful response.

module_id: String, required. Identifier of the module instance. The value is returned as a UUID.
module_class: String, required. Identifier of the module class. For this endpoint, the expected module class is assistant.
cid: String, required. Concatenation of module_id and module_class in the <module_id>@<module_class> format.
kind: String, required. Category of modules. For Assistant modules, the expected category is assistants.
name: String, required. Name of the module instance.
description: String, not required. Description of the module instance.
is_active: Boolean, required. Defines whether the module instance is active.
configuration: Object, required. Internal Assistant module configuration.
created_date: String, required. Date and time when the module instance was created.
updated_date: String, required. Date and time when the module instance was last updated.

Error responses

The endpoint can return the following error responses. Error response bodies are returned as strings with error details.

Status code	Status	Response body
400	Bad request	String with error details.
404	Not found	String with error details.
500	Internal error	String with error details.

Example request

The following example updates an Assistant module instance configuration.

Replace <api-endpoint>, {tenantId}, {projectId}, {moduleId}, and <token> with values from your environment.

curl -X PUT \
  "https://<api-endpoint>/assistant/v1/configuration/{tenantId}/{projectId}/modules/assistant/{moduleId}" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer <token>" \
  -d '{
    "configuration": {
      "agents": [
        "string"
      ],
      "generative_model_id": "mistralai/Mistral-Small-24B-Instruct-2501",
      "max_response_tokens": 1000,
      "stream": false,
      "system_instruction": "You are a helpful assistant.",
      "temperature": 1
    },
    "description": "Assistant module instance",
    "is_active": true,
    "name": "Assistant"
  }'

If the request succeeds, the API returns the updated module instance details with status code 200.

What to do next

After updating the module instance configuration, retrieve the module instance details or run a test conversation to verify that the new configuration is applied.

Check the saved assistant with a Playground conversation.
For console configuration, use the AI Assistants guide.

Still need help?

If this article doesn’t answer all of your questions, our support team will do their best to help you.

Contact support