Search results

tao-converter3.22.05 engine deploy failure

Status: Open.

#4343 In nvidia/TensorRT;

· oswanpy opened on Feb 1, 2025

after converting onnx fp32 to int8 engine with custom calibration, the engine layers still show fp32

Status: Open.

#4341 In nvidia/TensorRT;

· jinhonglu opened on Jan 28, 2025

Problems using pytorch stream

Status: Open.

#4340 In nvidia/TensorRT;

· ninono12345 opened on Jan 26, 2025

Unable to load engine in c++ api

Status: Open.

#4339 In nvidia/TensorRT;

· ninono12345 opened on Jan 26, 2025

Convert to fp16 results in NaN, polygraphy tests yield inconsistent results

AccuracyOutput mismatch between TensorRT and other frameworksOutput mismatch between TensorRT and other frameworks

triagedIssue has been triaged by maintainersIssue has been triaged by maintainers

Status: Open.

Bug

#4338 In nvidia/TensorRT;

· SunlightBoi opened on Jan 25, 2025

Onnx GroupNorm Op21 failed in TensorRT 10.7

internal-bug-trackedTracked internally, will be fixed in a future release.Tracked internally, will be fixed in a future release.

ONNXIssues relating to ONNX usage and importIssues relating to ONNX usage and import

triagedIssue has been triaged by maintainersIssue has been triaged by maintainers

Status: Open.

Bug

#4336 In nvidia/TensorRT;

· toothache opened on Jan 24, 2025

Error Code 1: Cask (Cask convolution execution) during inference with .to("cuda")

Status: Open.

#4335 In nvidia/TensorRT;

· yu0o0 opened on Jan 22, 2025

TensorRT 8.5.2 PASS but TensorRT 10.4.0 FAILED

Status: Open.

#4334 In nvidia/TensorRT;

· yijun02 opened on Jan 22, 2025

Comparison of infer speed between different TRT8.5.3.1 VS TRT 10.5.0.18 on GPU 3060-12G/4060Ti-16G

PerformanceGeneral performance issuesGeneral performance issues

Status: Open.

#4332 In nvidia/TensorRT;

· gaoyu-cao opened on Jan 21, 2025

TensorRT‌ 10.3 wrong results!

Status: Open.

#4330 In nvidia/TensorRT;

· OctaAIVision opened on Jan 19, 2025

support for libcudnn-9 and libcublas-12

Status: Open.

#4329 In nvidia/TensorRT;

· JohnRDOrazio opened on Jan 18, 2025

trt.IStreamReader (as implemented e.g. in polygraphy) requires higher peak CPU memory and more time than naive python implementation.

Status: Open.

#4327 In nvidia/TensorRT;

· michaelfeil opened on Jan 16, 2025