The Vitis AI runtime enables applications to use the unified high-level runtime API for both data center and Edge making data center-to-Edge deployments seamless and efficient.
Following are the features for the AI runtime API:
- Asynchronous submission of jobs to the accelerator
- Asynchronous collection of jobs from the accelerator
- C++ and Python implementations
- Support for multi-threading and multi-process execution
The framework of VART is shown in the following figure. For this Vitis AI release, VART is based on XRT. XIR is the AMD Intermediate Representation.
Figure 1. VART Stack