ND4J slow on M3

Hi,
i have imported nd4j-native:1.0.0-M2.1 via maven to my project. It’s running on m3 apple silicon.
When I run mmul operation in my code, it seems to be very slow compared to traditional implementation.

Is M3 supported or am i doing something wrong?

@pho001 can you file an issue with what you’re seeing? We just delegate to openblas on there so it’s probably good to check if there’s any issues on Me with openblas.