Can you have a try to set batch-size to the max number of the source?
https://docs.nvidia.com/metropolis/deepstream/dev-guide/text/DS_plugin_gst-nvinfer.html#id2
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| When I ran the deepstream example, I had to run the engine transformation on each execution | 3 | 148 | June 25, 2024 | |
| What changes to inference configuration necessitate a new .engine file | 8 | 60 | July 31, 2025 | |
| Error reported after replacing model in deepstream-test3 with yolov5 model | 5 | 473 | October 14, 2022 | |
| Is there anything that needs extra attention when using my own engine file? | 18 | 1423 | July 20, 2021 | |
| Batch-size with 9 rtsp streams | 16 | 1712 | October 12, 2021 | |
| Reshaping error when set batch-size greater than 1 in onnx modle | 23 | 1268 | February 10, 2023 | |
| Deepstream engine files not generated | 13 | 4739 | October 12, 2021 | |
| Unable to generate and Load Engine files on deepstream-7.1 | 46 | 114 | July 14, 2025 | |
| In deepstream-test3 use yolov7, only one source can run, >= 2 sources is error | 4 | 320 | April 19, 2023 | |
| TRT engine creation on source addition/deletion | 4 | 22 | July 18, 2025 |