> ## Documentation Index > Fetch the complete documentation index at: https://docs.epigos.ai/llms.txt > Use this file to discover all available pages before exploring further. # Object detection > Run inference on object detection models hosted on Epigos AI. Once your models are deployed on the Epigos AI platform, they come with a readily accessible REST API. ## OpenAPI ````yaml post /predict/detect/{model_id}/ openapi: 3.0.0 info: title: Epigos API version: 0.1.0 contact: name: Epigos Support email: hello@epigos.ai url: https://epigos.ai/contact termsOfService: https://epigos.ai/terms license: name: MIT url: https://raw.githubusercontent.com/Epigos-AI/docs/main/LICENSE description: |- Epigos AI API reference documentation. Our API is still very much a work in product and subject to change. servers: - url: https://api.epigos.ai security: - APIKeyHeader: [] - APIKeyQuery: [] paths: /predict/detect/{model_id}/: post: tags: - Detection summary: Object detection description: > Run inference on object detection models hosted on Epigos AI. Once your models are deployed on the Epigos AI platform, they come with a readily accessible REST API. parameters: - name: model_id in: path required: true example: 5581982d-721c-4e3c-b863-da2d985364a2 description: >- The unique ID of your model deployed on Epigos AI. You can find it in the web dashboard by viewing the model details. schema: type: string format: uuid title: Model Id requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/ObjectDetectionIn' examples: Predict Image URL: value: image: https://example.org/image.jpg Predict Base64 encoded image: value: image: base64 Predict with confidence threshold: value: image: https://example.org/image.jpg confidence: 0.7 Predict with annotation options: value: image: https://example.org/image.jpg annotate: true stroke_width: 2 responses: '200': description: Successful Response content: application/json: schema: $ref: '#/components/schemas/ObjectDetectionResult' examples: Object Detection: value: image: base64 detections: - label: car confidence: 0.954 x: 281 'y': 187 width: 46 height: 36 - label: person confidence: 0.854 x: 21 'y': 87 width: 40 height: 30 labelsCount: - label: car total: 1 - label: person total: 1 '400': description: Validation Error content: application/json: schema: $ref: '#/components/schemas/HTTPValidationError' '401': description: Unauthorized Error content: application/json: schema: $ref: '#/components/schemas/HTTPUnauthorizedError' components: schemas: ObjectDetectionIn: description: >- You can POST a base64 encoded image directly to your model endpoint. Or you can pass a URL as the image parameter in the request body if image is already hosted elsewhere properties: image: type: string title: Image description: >- Image path to run detections on. It can be base64 encoded string or URL to the image. example: https://example.org/image.jpg confidence: type: number maximum: 1 minimum: 0 title: Confidence description: The confidence threshold used to filter out predictions nullable: false example: 0.7 annotate: type: boolean title: Annotate description: >- Indicate whether to draw bounding boxes on the image for the detected objects nullable: false default: true show_prob: type: boolean title: Show Probabilities description: Indicate whether to show prediction confidences on each bounding box nullable: true default: true stroke_width: type: number title: Stroke width maximum: 5 minimum: 1 description: Bounding box border width nullable: true additionalProperties: false type: object required: - image title: ObjectDetectionIn ObjectDetectionResult: description: > The API inference route hosted by us returns a JSON object that includes a predictions array. Each prediction within this array comprises the following attributes: properties: detections: items: $ref: '#/components/schemas/DetectedObject' type: array title: Detections description: collection of all detected objects for the prediction image: type: string title: Image description: Base64-encoded string of the annotated image. nullable: true labelsCount: items: $ref: '#/components/schemas/ObjectLabelCount' type: array title: Labels Count description: aggregate of count of all detected objects additionalProperties: false type: object title: ObjectDetectionResult HTTPValidationError: title: Bad Request description: Bad Request (HTTP 400). properties: type: type: string description: '`error`.' enum: - error example: error code: type: string description: '`bad_request`.' example: bad_request enum: - bad_request status: type: integer description: '`400`.' example: 400 enum: - 400 message: type: string description: >- Describes the fields that are missing or incorrectly formatted in the API request. example: Missing '****' field details: type: array description: >- A list of detail objects that further clarify the reason for the error. items: $ref: '#/components/schemas/ErrorDetail' HTTPUnauthorizedError: title: Unauthorized Error description: Unauthorized Error (HTTP 401) properties: type: type: string description: '`error`.' enum: - error example: error code: type: string description: '`unauthorized`.' example: unauthorized enum: - unauthorized status: type: integer description: '`401`.' example: 401 enum: - 401 message: type: string description: No valid API authentication found. example: Invalid or missing credentials enum: - Invalid or missing credentials details: type: array example: [] description: >- A list of detail objects that further clarify the reason for the error. items: $ref: '#/components/schemas/ErrorDetail' DetectedObject: properties: label: type: string title: Label description: the label of the detected object. confidence: type: number title: Confidence description: the model's confidence that the image contains the detected objects. x: type: number title: X description: X-coordinates of the bounding box of the detected object. 'y': type: number title: 'Y' description: Y-coordinates of the bounding box of the detected object. width: type: number title: Width description: Width of the bounding box of the detected object. height: type: number title: Height description: Height of the bounding box of the detected object. additionalProperties: false type: object title: DetectedObject ObjectLabelCount: properties: label: type: string title: Label description: the label of the detected object. total: type: number title: Total description: the total number of detections for the label. additionalProperties: false type: object title: ObjectLabelCount ErrorDetail: description: Additional detail about the part of a request body that caused an issue properties: location: type: string example: body description: The location where the error caused an issue. enum: - query - body - path - header type: type: string example: value_error.missing description: A unique identifier for the type of error that occurred. pointer: type: string example: /image description: >- The exact item for which the validation did not succeed. This is a JSON pointer for request bodies, while for query, path, and header parameters it is the name of the parameter. message: type: string example: field required description: A human readable message for this error detail. securitySchemes: APIKeyHeader: type: apiKey in: header name: X-API-Key description: Your API key (obtained via your workspace API settings page) APIKeyQuery: type: apiKey in: query name: api-key description: Your API key (obtained via your workspace API settings page) ````