DeepStream 9.1 – Missing libgstnvdsseimeta.so.1.0.0 on Jetson Orin Nano

Accelerated Computing Intelligent Video Analytics DeepStream SDK
,
1

I’m using DeepStream 9.1 on a Jetson Orin Nano. nvstreammux is unavailable because libnvdsgst_multistream.so fails to load:

libgstnvdsseimeta.so.1.0.0: cannot open shared object file: No such file or directory
No such element or plugin 'nvstreammux'

ldd on libnvdsgst_multistream.so confirms the missing dependency:

libgstnvdsseimeta.so.1.0.0 => not found

I also searched the filesystem and the DeepStream 9.1 SDK archive, but this library is not present.

Environment:

  • Jetson Orin Nano
  • DeepStream 9.1
  • CUDA 13.2

Could you please confirm which package provides libgstnvdsseimeta.so.1.0.0 and the correct way to install it for DeepStream 9.1?

Thanks.

6

Orin Nano does not support DeepStream 9.1 because JetPack 7.2 does not support it.
The Orin Nano supports up to JP-6.1 / DS-7.1.

CUDA 13.2 is part of JetPack 7.2. How did you install it? Refer to this table

Even though installing this .deb file didn’t allow DS-9.1 to run correctly on the Orin Nano, I still checked the source.

  sudo apt update
  sudo apt install nvidia-l4t-gstreamer
7

Could you please clarify what you mean by “Orin Nano does not support DeepStream 9.1”?

The current NVIDIA DeepStream 9.1 documentation clearly mentions support for Jetson Orin (Nano, NX, AGX) and states that the Jetson package is based on JetPack 7.2 GA (r39.2 BSP).

DS9.1 Release Note

Also, the original libgstnvdsseimeta.so.1.0.0 / nvstreammux issue is resolved by installing nvidia-l4t-gstreamer (39.2.0-20260601141651), and DeepStream 9.1 is now working correctly on the Orin Nano.

8

Sorry, this appears to be an issue with the DeepStream documentation; the compatibility table I mentioned earlier indicates that JP-7.2 does not support the Orin Nano.

This link suggests that the Orin Nano does support JP-7.2. And the link you provided also confirms this.

If the method above solves your problem, I will look into the documentation.

Thanks for your feedback.