Aaqib
@chrisella Can you provide the command you ran to create the archive file?
How did you install mms? From source or pip?
Can you provide more details about your multi-model-server setup?
@dalvlv Can you provide some more detail on which handler are you using? I am assuming when you say dynamic batch size you are talking about client size batching
@sharvil We do support [building from source](https://github.com/pytorch/serve/blob/master/docker/build_image.sh#L66-L82) for different versions of cuda. You can check out [this](https://github.com/pytorch/serve/tree/master/docker#create-torchserve-docker-image) for more details
@HamidShojanazeri @yurkoff-mv I was able to register both workflows in the [nmt transformers](https://github.com/pytorch/serve/tree/master/examples/Workflows/nmt_tranformers_pipeline) simultaneously and run inferences on both. They share the same model mar file but are registered as...
@ayushch3 It seems like an issue with loading the model file. Will try and reproduce this but in the meantime do you mind getting a shell into the container and...
@DPeled Thanks for pointing out the bug. Would you like to send out the fix as a PR?
You are right. I did a quick search and found a few more [1](https://github.com/pytorch/serve/blob/873edc50cc6059c47830a5d30bbbdbe1e1a3d9bb/test/pytest/test_handler.py#L11) [2](https://github.com/pytorch/serve/blob/7a69e4018f3d694891c1415e2ab01413452fb009/test/pytest/test_metrics_kf.py#L15) [3](https://github.com/pytorch/serve/blob/dd2d48525e10ca401092b4abad0792191f7469b9/ts/model_server.py#L95) [4](https://github.com/pytorch/serve/blob/5b96c5fb6201e459a5e8417a09984638332e1533/benchmarks/utils/gen_md_report.py#L26) [5](https://github.com/pytorch/serve/blob/04c873f31abf5345eba8c93e6856a621b8a9398d/test/pytest/test_utils.py#L12) [6](https://github.com/pytorch/serve/blob/7a69e4018f3d694891c1415e2ab01413452fb009/test/pytest/test_metrics_kf.py#L15) If you'd like to contribute, feel free to send a PR
@mossaab0 What version of TS are you using? Can you try building TS from source and let me know if it still fails. I suspect this is the same issue...