---
title: "Inference (Auxen-native)"
method: POST
path: "/v1/{deploymentId}/chat"
tags: ["Inference"]
---

# Inference (Auxen-native)

`POST /v1/{deploymentId}/chat`

## Request

- Base URL: `https://api.auxen.ai`
- URL: `https://api.auxen.ai/v1/{deploymentId}/chat`
- Auth: HTTP bearer

## Path parameters

- `deploymentId` string, required

## Request body

- object
  - `message` string, required
  - `system` string
  - `history` object[]
    - `role` 'user' | 'assistant' | 'system'
    - `content` string
  - `stream` boolean
  - `options` object
    - `temperature` number
    - `num_predict` integer

## Response `200`

OK (non-streaming)

- object
  - `success` true, required
  - `data` object, required
    - `message` string
    - `model` string
    - `usage` object
      - `prompt_tokens` integer
      - `completion_tokens` integer
      - `total_tokens` integer
    - `response_time_ms` integer
  - `meta` Meta, required
    - `requestId` string
    - `timestamp` string, date-time

## Other responses

- `401`: Missing or invalid API key.
- `409`: Deployment not ready yet — poll until status is ready.

---

[API](https://skmtc.dev/auxen/apis/auxen-headless-api.md) · [All operations](https://skmtc.dev/auxen/apis/auxen-headless-api/llms.txt) · [OpenAPI document](https://skmtc.dev/auxen/apis/auxen-headless-api/revisions/164be65f5ce8?raw)