NVidia TensorRT: high-performance deep learning inference accelerator (TensorFlow Meets)

In this episode of TensorFlow Meets, we are joined by Chris Gottbrath from NVidia and X.Q. from the Google Brain team to talk about NVidia TensorRT. NVidia TensorRT is a high-performance, programmable inference accelerator that delivers low latency and high-throughput for deep learning applications. Developers can create neural networks and artificial intelligence to run their networks in productions or devices with the full performance that GPUs can offer. NVidia TensorRT allows developers to enjoy the diversity and flexibility of TensorFlow, while still having the high accuracy and performance that TensorRT provides. Watch to learn more and leave your questions for the TensorFlow team in the comments below!

[X_platform]

Would you consider stepping through a couple of example codes in future interview like this? Much appreciated

[oscarriddleus]

I just met with Laurance in TF roadshow 2018 in Beijing, and I also once met with Chris in my office. Here I found both them two in my recommended video, that's interesting~

[sibyjoseplathottam]

Does TensorRT work with GeForce GPU's?

[Benimation]

When will it run op OpenCL? Or maybe even Metal?

[제갈식]

Hi, Trained model loading is so slow. So, Case of One image inference,
CPU is faster than CUDA. How can I fast inference with CNN?

[kkeshaw]

damm this is next gen real shit i will definatlly dive into it :)