Multi-model Cascading Inference

The basic process of multi-model inference is similar to that of single-model inference. For details, see Model Inference with Static-Shape Inputs.

The differences between multi-model inference and single-model inference in the usage of the ACL API are as follows:

• If you need to load more models, call the model loading API separately for each model. For details about how to load a model, see Loading a Model.
• If you need to run more models, call the model execution API separately for each model. For details about how to run a model, see Running a Model.