---
title: "LyticsSegModel: Create job"
method: POST
path: "/job/lytics-seg-model"
tags: ["Jobs"]
---

# LyticsSegModel: Create job

`POST /job/lytics-seg-model`

## Request

- Base URL: `https://api.lytics.io/v2`
- URL: `https://api.lytics.io/v2/job/lytics-seg-model`
- Auth: API key in header `Authorization`

Create a LyticsSegModel job

## Query parameters

- `account_id` string

## Request body `*/*`

- object
  - `auth_ids` string[]: Authorization IDs associated with the Job
  - `config` object, required: Job configuration, unique per-workflow
    - `additional_fields` string[]: Additional fields to use as model features
    - `auto_tune` boolean: Use [Lytics Auto Tune](https://learn.lytics.com/documentation/product/features/laboratory/lookalike-models/interpreting-model-messages#improving-unhealthy-models) to expand feature candidates to all available user fields
    - `blocked_fields` string[]: Blocked fields to omit as model features
    - `custom_segment_ids` string[]: Select any segments to add to the model as features
    - `hidden` boolean: Hide the model from the UI
    - `max_rounds` integer: "The number of rounds to iterate on the model (only for GBM types)"
    - `model_name` string: Enter a custom model name; if no name is provided, the source and target slug names will be used
    - `model_only` boolean: Check to build a sample model without deploying to all users
    - `model_type` string: Algorithm to use for model training
    - `num_to_train` integer: Size of the sample to use for model training
    - `re_run` boolean: Check to continuously re-train the model
    - `source` string: Choose the audience to use for the source (seed) segment
    - `tags` string[]: Tags to apply to a model for organizational purposes
    - `target` string: Choose the audience to use for the target segment
    - `use_content` boolean: Check to use content affinity in the model
    - `use_scores` boolean: Check to use behavioral scores in the model
  - `description` string: A user-defined description
  - `drop_events_during_quiet_window` boolean
  - `expires_at` string, date-time
  - `monitoring` object: Inline monitoring config. On write: omit to keep, null to clear, or a config to set.
  - `name` string, required: A user-defined name
  - `quiet_period` integer
  - `quiet_time_of_day` string
  - `quiet_timezone` string

## Response `201`

LyticsSegModel Job

- object
  - `_meta` object: Response Metadata
  - `data` object: Response Payload
    - `account_id` string: Lytics Account ID
    - `anomaly_rule` ModelsAnomalyRuleResp
      - `aid` integer
      - `checks` ModelsRuleCheckConfigReq[]
        - `allowable_violations` integer
        - `interval` string
        - `lookback_count` integer
        - `max_threshold` integer
        - `metric_type` string
        - `min_threshold` integer
        - `type` string
        - `variance_pct` number
      - `created` string
      - `id` string
      - `subject_id` string
      - `subject_type` string
      - `updated` string
    - `api_token_auth_id` string: this is set when work is create using access token
    - `auth_ids` string[]: Authorization IDs associated with the Job
    - `config` object, required: Job configuration, unique per-workflow
      - `additional_fields` string[]: Additional fields to use as model features
      - `auto_tune` boolean: Use [Lytics Auto Tune](https://learn.lytics.com/documentation/product/features/laboratory/lookalike-models/interpreting-model-messages#improving-unhealthy-models) to expand feature candidates to all available user fields
      - `blocked_fields` string[]: Blocked fields to omit as model features
      - `custom_segment_ids` string[]: Select any segments to add to the model as features
      - `hidden` boolean: Hide the model from the UI
      - `max_rounds` integer: "The number of rounds to iterate on the model (only for GBM types)"
      - `model_name` string: Enter a custom model name; if no name is provided, the source and target slug names will be used
      - `model_only` boolean: Check to build a sample model without deploying to all users
      - `model_type` string: Algorithm to use for model training
      - `num_to_train` integer: Size of the sample to use for model training
      - `re_run` boolean: Check to continuously re-train the model
      - `source` string: Choose the audience to use for the source (seed) segment
      - `tags` string[]: Tags to apply to a model for organizational purposes
      - `target` string: Choose the audience to use for the target segment
      - `use_content` boolean: Check to use content affinity in the model
      - `use_scores` boolean: Check to use behavioral scores in the model
    - `created` string
    - `deleted` boolean: The Job has been deleted
    - `description` string: A user-defined description
    - `drop_events_during_quiet_window` boolean
    - `expires_at` string, date-time
    - `id` string: A Lytics-assigned, immutable unique ID for the Job
    - `meta` object: Extra meta data about the job
    - `monitoring` object: Inline monitoring config. On write: omit to keep, null to clear, or a config to set.
    - `name` string, required: A user-defined name
    - `quiet_period` integer
    - `quiet_time_of_day` string
    - `quiet_timezone` string
    - `sleep_until` string
    - `status` 'runnable' | 'sleeping' | 'paused' | 'completed' | 'fault' | 'failed' | 'killed' | 'deleted' | 'unknown': The status of the job
    - `updated` string
    - `user_id` string: The user ID that created or updated the job
    - `workflow` string: The workflow slug
    - `workflow_id` string: The workflow ID
  - `request_id` string
  - `status` integer: HTTP Status Code

## Other responses

- `404`: Not Found
- `500`: Internal Server Error

## Changes

- **2026-11-03** `d8e975057aec`: 1 info
  - endpoint added

[Operation history](https://skmtc.dev/lytics/apis/lytics-api/changes/job/lytics-seg-model/post.md)

---

[API](https://skmtc.dev/lytics/apis/lytics-api.md) - [All operations](https://skmtc.dev/lytics/apis/lytics-api/llms.txt) - [OpenAPI document](https://skmtc.dev/lytics/apis/lytics-api/revisions/d8e975057aec?raw)