---
title: "Create Endpoint"
method: POST
path: "/api/v2/inference/endpoints"
tags: ["Serverless (compat)"]
---

# Create Endpoint

`POST /api/v2/inference/endpoints`

## Request body

- ServerlessEndpointCreate
  - `name` string
  - `mode` string
  - `managed_engine` string
  - `model_name` string
  - `model_ref` string
  - `image_ref` string
  - `docker_image` string
  - `source_type` string
  - `source_ref` string
  - `source_ref_branch` string
  - `gpu_type` string
  - `gpu_tier` string
  - `gpu_count` integer
  - `region` string
  - `min_workers` integer
  - `max_workers` integer
  - `max_concurrency` integer
  - `idle_timeout_sec` integer
  - `scaledown_window_sec` integer, nullable
  - `scaling_policy_type` string
  - `scaling_policy_value` integer
  - `execution_mode` string
  - `queue_timeout_sec` integer
  - `startup_command` string
  - `http_port` integer
  - `health_check_path` string
  - `health_endpoint` string
  - `cuda_version` string
  - `request_timeout_sec` integer
  - `max_request_bytes` integer
  - `env` object
  - `lora_adapters` object[]

## Response `200`

Successful Response

- unknown

## Other responses

- `422` — Validation Error

---

[API](https://skmtc.net/aabiro/apis/xcelsior.md) · [All operations](https://skmtc.net/aabiro/apis/xcelsior/llms.txt) · [OpenAPI document](https://skmtc-service-staging.skmtc.workers.dev/v1/apis/aabiro/xcelsior/versions/b4f5ef136e6a/schema)
