Labels · NVIDIA/TensorRT

  • This issue or pull request already exists

  • Request for new functionality

  • Tracked internally, will be fixed in a future release.

  • Issue is under investigation by TensorRT devs

  • Output mismatch between TensorRT and other frameworks

  • Issues regarding demos under the demo/ directory: Diffusion, DeBERTa, Bert

  • Lack of clarity or potential improvements for docs

  • issues when using TensorRT on embedded platforms

  • Issues with building TensorRT engines

  • Issues with ONNX-Graphsurgeon

  • Issues relating to ONNX usage and import

  • Building issues: cmake, make, pip install, etc.

  • General performance issues

  • Issues when using TensorRT plugins

  • Issues related to Quantization

  • Other generic runtime issues that does not fall into other modules

  • Issues when using TensorRT samples under the samples/ directory, including usage with trtexec

  • PR or issue hasn't been responded to by the OP

  • Issue has been triaged by maintainers

  • Requires more information from author of item to make progress on the issue.

  • This will not be worked on