---
title: "Create Serving Endpoint"
method: POST
path: "/v1/serving/endpoints"
tags: ["serving"]
---

# Create Serving Endpoint

`POST /v1/serving/endpoints`

## Request body

- OnlineServingEndpointRequest
  - `model_id` union, required
    - string, uuid
    - string
  - `predict_options` OnlinePredictionOptions
    - `binary_classification_threshold` number, nullable — Probability cutoff used to derive a predicted class for binary classification models. Null leaves predictions as raw scores.
    - `num_classes_to_return` integer, nullable — For multiclass models, cap the number of top class probabilities returned per row. Null returns all.
  - `max_qps` integer — Advisory per-endpoint request-rate ceiling.

## Response `201`

Successful Response

- OnlineServingEndpointResource
  - `id` union, required
    - string, uuid
    - string
  - `endpoint_url` string, required — Public base URL of the endpoint, e.g. https://ep-{id}.serving.<zone>.
  - `config` OnlineServingEndpointRequest, required
    - `model_id` union, required
      - string, uuid
      - string
    - `predict_options` OnlinePredictionOptions
      - `binary_classification_threshold` number, nullable — Probability cutoff used to derive a predicted class for binary classification models. Null leaves predictions as raw scores.
      - `num_classes_to_return` integer, nullable — For multiclass models, cap the number of top class probabilities returned per row. Null returns all.
    - `max_qps` integer — Advisory per-endpoint request-rate ceiling.
  - `launched_at` string, date-time, required
  - `status` OnlineServingStatus, required
    - `status_code` 'ready' | 'dormant' | 'in_progress' | 'failed', required
    - `last_updated_at` string, date-time, required
    - `stage` string, nullable — Human-readable sub-step while ``in_progress`` (e.g. provisioning_routing).
    - `failure_message` string, nullable — Populated when ``status_code == 'failed'``.

## Other responses

- `422` — Validation Error

---

[API](https://skmtc.dev/outerproduct/apis/outerproduct-api.md) · [All operations](https://skmtc.dev/outerproduct/apis/outerproduct-api/llms.txt) · [OpenAPI document](https://skmtc-service-production.skmtc.workers.dev/v1/apis/outerproduct/outerproduct-api/revisions/a49cdafe7fba/schema)