Bernard Ryhede Bengtsson

FP16 Accuracy failure of TensorRT 8.6.3 when running trtexec built engine on GPU RTX4090

7

## Description We are trying to recreate the results from: [https://arxiv.org/abs/2402.05008](url). Using the same .onnx file to compile engines provided by the authors, we find that using fp16 model has...

Benchmarking inference of TensorRT 8.6.3 using trtexec on GPU RTX 4090

3

## Description We're benchmarking my mixed-precision models using: `trtexec --loadEngine=model.engine --useCudaGraph --iterations=100 --avgRuns=100` We compared two models, one baseline in FP16, and another model where we reduced the precision for...

Discrepancy between TRT engines for same models - TensorRT Issue cross-reference

1

We are trying to recreate the results for EfficientViT-SAM, but are having issues with certain models. In this case, we reported issues with l2_encoder.onnx. In summary, these are the results:...