{
  "opencollection": "1.0.0",
  "info": {
    "name": "Triton Inference Server NVIDIA Triton Inference Server HTTP/REST CUDA Shared Memory Model Repository API",
    "version": "2.0"
  },
  "items": [
    {
      "info": {
        "name": "Model Repository",
        "type": "folder"
      },
      "items": [
        {
          "info": {
            "name": "Triton Inference Server List Models in the Repository",
            "type": "http"
          },
          "http": {
            "method": "POST",
            "url": "http://localhost:8000/v2/repository/index",
            "body": {
              "type": "json",
              "data": "{}"
            }
          },
          "docs": "Retrieve the index of all models available in the model repository. Can optionally filter to show only models that are ready. This is a Triton extension to the KServe protocol."
        },
        {
          "info": {
            "name": "Triton Inference Server Load or Reload a Model",
            "type": "http"
          },
          "http": {
            "method": "POST",
            "url": "http://localhost:8000/v2/repository/models/:model_name/load",
            "params": [
              {
                "name": "model_name",
                "value": "",
                "type": "path",
                "description": "Name of the model"
              }
            ],
            "body": {
              "type": "json",
              "data": "{}"
            }
          },
          "docs": "Request that a model be loaded into Triton, or reloaded if it is already loaded. Optionally provide model configuration overrides as parameters. This is a Triton extension to the KServe protocol."
        },
        {
          "info": {
            "name": "Triton Inference Server Unload a Model",
            "type": "http"
          },
          "http": {
            "method": "POST",
            "url": "http://localhost:8000/v2/repository/models/:model_name/unload",
            "params": [
              {
                "name": "model_name",
                "value": "",
                "type": "path",
                "description": "Name of the model"
              }
            ],
            "body": {
              "type": "json",
              "data": "{}"
            }
          },
          "docs": "Request that a model be unloaded from Triton. Once unloaded the model will no longer be available for inference. This is a Triton extension to the KServe protocol."
        }
      ]
    }
  ],
  "bundled": true
}