---
title: "Create RunningSession"
method: POST
path: "/trained_models/{trained_model_id}/running_sessions"
---

# Create RunningSession

`POST /trained_models/{trained_model_id}/running_sessions`

## Request

- Base URL: `https://darwin.v7labs.com/ai`
- URL: `https://darwin.v7labs.com/ai/trained_models/{trained_model_id}/running_sessions`
- Auth: API key in header `Authorization`

Create a RunningSession to elastically run a TrainedModel

## Path parameters

- `trained_model_id` string, required

## Request body

- object
  - `min` integer: The minimum number of instances to run for the TrainedModel
  - `max` integer: The maximum number of instances to run for the given TrainedModel, should the model need to scale up
  - `device` string: Either "cpu" or "gpu", provided that the ModelTemplate associated to the TrainedModel includes it within its list of supported devices

## Response `200`

200

- object
  - `access_level` string
  - `device` string
  - `id` string
  - `inserted_at` string
  - `max` integer
  - `meta` object
  - `min` integer
  - `name` string
  - `team_id` integer
  - `trained_model_id` string
  - `updated_at` string

---

[API](https://skmtc.dev/v7labs/apis/wind.md) - [All operations](https://skmtc.dev/v7labs/apis/wind/llms.txt) - [OpenAPI document](https://skmtc.dev/v7labs/apis/wind/revisions/4035e136c087?raw)