---
title: "Per-model runtime states"
method: GET
path: "/v1/models/state"
tags: ["Engine"]
---

# Per-model runtime states

`GET /v1/models/state`

## Request

- Base URL: `http://localhost:9998`
- URL: `http://localhost:9998/v1/models/state`
- Auth: HTTP bearer

Returns the runtime state of every tracked model (cold, loading, loaded,
inferring, unloading, error).

## Response `200`

Model state snapshot

- object
  - `states` ModelState[], required
    - `modelId` string, required
    - `state` 'cold' | 'loading' | 'loaded' | 'inferring' | 'unloading' | 'error', required
    - `backendId` string, required
    - `inferenceCount` number, required
    - `timestamp` string, date-time, required
    - `error` string

---

[API](https://skmtc.dev/modelpiper/apis/toolpiper-api.md) · [All operations](https://skmtc.dev/modelpiper/apis/toolpiper-api/llms.txt) · [OpenAPI document](https://skmtc.dev/modelpiper/apis/toolpiper-api/revisions/6722bf56385c?raw)