Nan

> Hi, @NicolasHug, Would you mind telling where do you get this information? i could not find it in the cuda 11.6 [release note](https://docs.nvidia.com/cuda/cuda-toolkit-release-notes/index.html). And I cannot reproduce this memory...

> > Would you mind telling where do you get this information? > > I basically tried all versions I could find from https://pkgs.org/search/?q=libnvjpeg-devel It seems that there is a...

> > I tried both on GPU 1070 and 3090 with the same CUDA version 11.7, but it only works on the 3090. So, I think cv-cuda is not compatible...

Only contiguous torch.Tensor is supported _at this moment_: `cvcuda.as_tensor(imgs.contiguous(), "NHWC") `

建模深度学习的部署用的， 主要是处理多级结构化，长尾需求，低代码化

> Mixtral MoE was merged last week [microsoft/onnxruntime#19945](https://github.com/microsoft/onnxruntime/pull/19945) Hi, @mgiessing , do you know what's the proper way to get ONNX from PyTorch for the MoE model? torch.onnx.export might involve...

Hi, guys, is there any update on this issue? The throughout is really high.

Use GPU preprocess and pipeline parallelism if you can. for instance, DALI, CVCUDA, [tp](https://torchpipe.github.io/docs/backend-reference/torch)

> > Could you please try latest 9.2, IIRC we add support to opset 17 since TRT 8.6. > > Download from https://developer.nvidia.com/downloads/compute/machine-learning/tensorrt/9.2.0/tensorrt-9.2.0.5.linux.x86_64-gnu.cuda-11.8.tar.gz https://developer.nvidia.com/downloads/compute/machine-learning/tensorrt/9.2.0/tensorrt-9.2.0.5.linux.x86_64-gnu.cuda-12.2.tar.gz https://developer.nvidia.com/downloads/compute/machine-learning/tensorrt/9.2.0/tensorrt-9.2.0.5.ubuntu-22.04.aarch64-gnu.cuda-12.2.tar.gz > > Hi @zerollzeng, thank...

can you share your python and c++ code?