models

The models list defines the AI inference pipeline. It can contain detectors, classifiers, and an optional tracker. Entries are processed in the order they appear, which controls the inference chain.

models:
  - name: "my-detector"
    type: "detector"
    architecture: "yolo"
    onnx_file: /path/to/model.onnx
    label_file: /path/to/labels.txt

Model fields

Field	Type	Required	Default	Description
name	string	No	auto-generated	Unique name for this model. Used by operate_on for chaining.
type	string	Yes	—	detector, classifier, or tracker
architecture	string	Yes (detectors)	—	Model architecture (e.g., yolo)
onnx_file	string	Yes	—	Path to ONNX model file
label_file	string	Yes	—	Path to label file (one class name per line)
batch_size	integer	No	next power of 2 ≥ stream count	Inference batch size
precision	string	No	"fp16"	Inference precision: fp32 or fp16
active_classes	list of int	No	all classes	Only detect these class IDs (detectors only)
operate_on	string	Yes (classifiers)	—	Name of the parent model to chain from
operate_on_classes	list of int	No	all parent classes	Only classify objects of these class IDs from the parent model

tip

The num_classes is automatically determined by counting lines in label_file. The DeepStream nvinfer config file is auto-generated — you only need to provide the ONNX model and labels.

See Model Types for detailed usage of each type.