Exports

Converts a trained model to another deployment format.

post/api/models/{owner}/{project}/{model}/exports

Request

  • Base URL
    https://platform.ultralytics.com
  • Auth
    HTTP bearer

Path parameters

ownerstring required
Example:ultralytics

Project owner

projectstring required
Example:yolo26

Project name

modelstring required
Example:yolo26n

Model name

Request body

format'torchscript' | 'onnx' | 'openvino' | 'engine' | 'coreml' | 'coreai' | 'saved_model' | 'pb' | 'edgetpu' | 'litert' | 'paddle' | 'mnn' | 'ncnn' | 'imx' | 'rknn' | 'executorch' | 'axelera' | 'deepx' | 'qnn' | 'hailo' | 'ascend' | 'xilinx' | 'nxp' required

Target export format

gpuType'rtx-2000-ada' | 'rtx-a4500' | 'rtx-a5000' | 'rtx-4000-ada' | 'l4' | 'a40' | 'rtx-3090' | 'rtx-a6000' | 'rtx-pro-4000' | 'rtx-pro-4500' | 'rtx-4090' | 'rtx-6000-ada' | 'l40s' | 'rtx-pro-5000' | 'rtx-5090' | 'l40' | 'a100-80gb-pcie' | 'a100-80gb-sxm' | 'rtx-pro-6000' | 'h100-pcie' | 'h100-nvl' | 'h100-sxm' | 'h200-nvl' | 'h200-sxm' | 'b200' | 'b300' | 'jetson-thor-t5000' | 'jetson-thor-t4000' | 'jetson-agx-orin-64gb' | 'jetson-agx-orin-32gb' | 'jetson-orin-nx-16gb' | 'jetson-orin-nx-8gb' | 'jetson-orin-nano-8gb' | 'jetson-orin-nano-4gb'

Target GPU type for TensorRT exports

Response

Export started

idstring required

Export ID

format'torchscript' | 'onnx' | 'openvino' | 'engine' | 'coreml' | 'coreai' | 'saved_model' | 'pb' | 'edgetpu' | 'litert' | 'paddle' | 'mnn' | 'ncnn' | 'imx' | 'rknn' | 'executorch' | 'axelera' | 'deepx' | 'qnn' | 'hailo' | 'ascend' | 'xilinx' | 'nxp' required

Model export format

status'queued' | 'running' required
gpuTypestring
region'us' | 'eu' | 'ap' required

Data region where data is stored

Changes

    • warning

      added the new nxp enum value to the response property for the response status

    • info

      added the new nxp enum value to the request property

    • breaking

      the // request property's max was decreased to 4096.00