---
title: "Detect Objects"
method: POST
path: "/v1/vision/detect"
tags: ["vision", "vision-detection"]
---

# Detect Objects

`POST /v1/vision/detect`

## Request

- The document declares no server URL.
- Auth: none declared
- TypeScript: <https://skmtc.dev/llama-farm/apis/llamafarm-edge-runtime/docs/v1/vision/detect/post.ts> (a function with typed inputs; its response is checked against the schema)

Detect objects in an image using YOLO.

## Request body

- DetectRequest
  - `image` string, required: Base64-encoded image
  - `model` string
  - `confidence_threshold` number
  - `classes` string[], nullable

## Response `200`

Successful Response

- DetectResponse
  - `detections` Detection[], required
    - `box` BoundingBox, required
      - `x1` number, required
      - `y1` number, required
      - `x2` number, required
      - `y2` number, required
    - `class_name` string, required
    - `class_id` integer, required
    - `confidence` number, required
  - `model` string, required
  - `inference_time_ms` number, required

## Other responses

- `422`: Validation Error

---

[API](https://skmtc.dev/llama-farm/apis/llamafarm-edge-runtime.md) - [All operations](https://skmtc.dev/llama-farm/apis/llamafarm-edge-runtime/llms.txt) - [OpenAPI document](https://skmtc.dev/llama-farm/apis/llamafarm-edge-runtime/revisions/a23a9e72958b?raw)