Release date

November 20th, 2020

Major features and improvements

New Compiler

Inference Engine

Benchmark Test


The following bugs are fixed:

1) Can Not Convert Yolo.h5 To Openvino Runtime.

2) gRPC:Received message larger than max.

3) Return Message Is Wrong When cudaMalloc() Failed In initializeOutputBindings() Method.

4) Can Not Do Predict With Following Transferred YoloV3 Model.

5) adlik_serving --help should exit successfully.

6) benchmark cant auto infer by tensorflow gpu image.

7) Prediction will fail if information in model.pbtxt and model representation not consistent in tensorflowLite runtime.

Please see for more information.