All issues

Search results

W4A16 FP4 build succeeds while silently realizing FP32 (no warning, no refusal)

Module:QuantizationIssues related to QuantizationIssues related to Quantization

Status: Open.

#4833 In NVIDIA/TensorRT;

· geoffrey-delhomme opened on Aug 14, 2026

Myelin "Could not infer output types for operation: dequantize" on SM 8.7 (Orin); identical engine builds on SM 11.0 (Thor) with the same TensorRT 10.16.2.10

Module:QuantizationIssues related to QuantizationIssues related to Quantization

Status: Open.

#4832 In NVIDIA/TensorRT;

· geoffrey-delhomme opened on Aug 14, 2026

[Architecture Discussion] Overhead and safety of asynchronous soft-stop signaling for KV-cache reclamation during generation loops

Feature RequestRequest for new functionalityRequest for new functionality

Status: Open.

#4831 In NVIDIA/TensorRT;

· alexbuiko-sketch opened on Aug 14, 2026

Myelin NVRTC compilation failure on Jetson Thor (SM110) building Fast-FoundationStereo - reproduces on TRT 10.14.1 and 11.2.1.2

Module:Embeddedissues when using TensorRT on embedded platformsissues when using TensorRT on embedded platforms

Module:Engine BuildIssues with building TensorRT enginesIssues with building TensorRT engines

Status: Open.

#4830 In NVIDIA/TensorRT;

· juliusekela opened on Aug 13, 2026

TensorRT 11.0.0.114 builder finds no valid tactic for BF16 ConvTranspose2d (works fine in FP16)

Module:Engine BuildIssues with building TensorRT enginesIssues with building TensorRT engines

Status: Open.

#4829 In NVIDIA/TensorRT;

· robbysun opened on Aug 12, 2026

Deprecate demoDiffusion

Module:DemoIssues regarding demos under the demo/ directory: Diffusion, DeBERTa, BertIssues regarding demos under the demo/ directory: Diffusion, DeBERTa, Bert

Status: Open.

#4827 In NVIDIA/TensorRT;

· pkisfaludi-nv opened on Aug 11, 2026

TRT 11.2: Transpose of a network input folded into a fused MHA operand with the permutation applied to the shape but not to the strides

Module:Engine BuildIssues with building TensorRT enginesIssues with building TensorRT engines

Status: Open.

#4826 In NVIDIA/TensorRT;

· vient opened on Aug 10, 2026

Wrong encoder output for NeMo Parakeet (FastConformer) on RTX 5090 (sm_120a) — all precisions, all opt levels, TRT 10.14 and 11.1

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

Status: Open.

#4822 In NVIDIA/TensorRT;

· cribbinsb opened on Jul 23, 2026

AveragePool fails at the exact batch boundary N=65536 on RTX 5090

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

Status: Open.

#4821 In NVIDIA/TensorRT;

· Alcoholrithm opened on Jul 22, 2026

YOLOv10 Error with TensorRT 10.13

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

Status: Open.

#4820 In NVIDIA/TensorRT;

· BobHo5474 opened on Jul 20, 2026

TRT 11.1 strongly-typed FP32 engine silently produces wrong results for a DETR-style detection model (correct in ONNXRuntime and TRT 10.13)

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

Status: Open.

#4813 In NVIDIA/TensorRT;

· ArgoHA opened on Jul 1, 2026

ResizeWithPadPlugin — Aspect-Ratio-Preserving Resize with Padding

Feature RequestRequest for new functionalityRequest for new functionality

Status: Open.

#4811 In NVIDIA/TensorRT;

· abdulrahman-hassanin opened on Jun 28, 2026