Release Notes — NVIDIA TensorRT Documentation (original) (raw)

These Release Note describe the key features, software enhancements and improvements, and known issues for the TensorRT release product package.

To review the TensorRT 10.7.0 and earlier documentation, refer to the TensorRT Archived Documentation.

To review the TensorRT 10.8.0 and later documentation, choose a version from the bottom left navigation selector toggle.

TensorRT 10.10.0#

These are the TensorRT 10.10.0 Release Notes, which apply to x86 Linux and Windows users, and Arm-based CPU cores for Server Base System Architecture (SBSA) users on Linux. This release includes several fixes from the previous TensorRT releases and additional changes.

Key Features and Enhancements

This TensorRT release includes the following key features and enhancements.

• TensorRT 10.10 enhances support for large tensors; with most layers now capable of handling large volumes. For more information about each layer’s capabilities, refer to the TensorRT Operator documentation.
• Improved performance for Tanh, Slice, and Concatenate operations on NVIDIA Blackwell GPUs.
• Improved performance for BF16 and FP16 batched GEMMs with static shapes and with small M/N/K dimensions (<=64) on NVIDIA Hopper and Blackwell GPUs.
• Improved MHA performance and enhanced mask support for MHA on Blackwell GPUs.
• Enabled GEMM+SwiGLU/GeGLU fusion on Blackwell GPUs for FP8 and FP16 data types.
• Improved FP8 and BF16 MHA performance on Ada GPUs for long sequence lengths.
• Improved MHA fusion logic to detect wider variants of MHA patterns. Refer to the Multi-Head Attention Fusion section for more information.
• Improved ConvNets performance in INT8 precision on Blackwell GPUs.
• Added PreviewFeature::kRUNTIME_ACTIVATION_RESIZE_10_10 which allows TensorRT to reduce memory usage by using better estimation of activation memory requirements based on actual input shapes during runtime.
• Added IPluginRegistry::getAllCreatorsRecursive() which enables TensorRT plugins to obtain a list of all plugin creators registered in TensorRT’s plugin registry.
• Improved performance for TensorRT builder resource initialization on pre-Blackwell platforms.

Compatibility

• TensorRT 10.10.0 has been tested with the following:
  • TensorFlow 2.13.1
  • PyTorch >= 2.0 (refer to the requirements.txt file for each sample)
  • ONNX 1.16.0
• This TensorRT release supports NVIDIA CUDA:
• This TensorRT release requires at least NVIDIA driver r450 on Linux or r452 on Windows as required by CUDA 11.0, which is the minimum CUDA version supported by this TensorRT release. For CUDA 12.x the minimum NVIDIA driver version is r535.

Limitations

• In some rare cases, FP8 MHA on SM90 might have accuracy issues with sequence length < 256.
• There are no optimized FP8 Convolutions for Group Convolutions and Depthwise Convolutions. Therefore, INT8 is still recommended for ConvNets containing these convolution ops.
• The FP8 Convolutions only support input/output channels, which are multiples of 16. Otherwise, TensorRT will fall back to non-FP8 convolutions.
• The FP8 Convolutions do not support kernel sizes larger than 32, such as 7x7 convolutions, and FP16 or FP32 fallback kernels will be used with suboptimal performance. Therefore, do not add FP8 Q/DQ ops before Convolutions with large kernel sizes for better performance.
• There cannot be any pointwise operations between the first batched GEMM and the softmax inside FP8 MHAs, such as having an attention mask. This will be improved in future TensorRT releases.
• On QNX, networks that are segmented into a large number of DLA loadables may fail during inference.
• The DLA compiler can remove identity transposes but cannot fuse multiple adjacent transpose layers into a single transpose layer (likewise for reshaping). For example, given a TensorRT IShuffleLayer consisting of two non-trivial transposes and an identity reshape in between, the shuffle layer is translated into two consecutive DLA transpose layers unless the user merges the transposes manually in the model definition in advance.
• nvinfer1::UnaryOperation::kROUND or nvinfer1::UnaryOperation::kSIGN operations of IUnaryLayer are not supported in the implicit batch mode.
• For networks containing normalization layers, particularly if deploying with mixed precision, target the latest ONNX opset containing the corresponding function ops, such as opset 17 for LayerNormalization or opset 18 GroupNormalization. Numerical accuracy using function ops is superior to the corresponding implementation with primitive ops for normalization layers.
• Weight streaming mainly supports GEMM-based networks like Transformers for now. Convolution-based networks may have only a few weights that can be streamed.
• When two convolutions with INT8-QDQ and residual add share the same weight, constant weight fusion will not occur. Make a copy of the shared weight for better performance.
• When building the nonZeroPlugin sample on Windows, you may need to modify the CUDA version specified in the BuildCustomizations paths in the vcxproj file to match the installed version of CUDA.
• The weights used in INT4 weights-only quantization (WoQ) cannot be refitted.
• The high-precision weights used in FP4 double quantization are not refittable.
• Python samples do not support Python 3.13. Only the 3.13 Python bindings are currently supported.

Deprecated API Lifetime

• APIs deprecated in TensorRT 10.10 will be retained until 4/2026.
• APIs deprecated in TensorRT 10.9 will be retained until 3/2026.
• APIs deprecated in TensorRT 10.8 will be retained until 2/2026.
• APIs deprecated in TensorRT 10.7 will be retained until 12/2025.
• APIs deprecated in TensorRT 10.6 will be retained until 11/2025.
• APIs deprecated in TensorRT 10.5 will be retained until 10/2025.
• APIs deprecated in TensorRT 10.4 will be retained until 9/2025.
• APIs deprecated in TensorRT 10.3 will be retained until 8/2025.
• APIs deprecated in TensorRT 10.2 will be retained until 7/2025.
• APIs deprecated in TensorRT 10.1 will be retained until 5/2025.

Refer to the API documentation (C++, Python) for instructions on updating your code to remove the use of deprecated features.

Deprecated and Removed Features

The following features have been deprecated or removed in TensorRT 10.10.0.

• Deprecated PluginVersion and PluginCreatorVersion enums. These are only used in relation to IPluginV2-descendent plugins and IPluginCreator-descendent plugin creators, respectively, which have all been deprecated as of 10.10.

Fixed Issues

• FP8 MHA performance may have been lower than BF16/FP16 MHA on SM89 when the sequence length was long (for example, >100k).
• The scale factor had to be a build-time constant if QuantizeLayer was used with the output FP4 data type.
• Fixed a memory leak issue when TensorRT builds the engine on pre-Blackwell GPUs.
• The Valgrind Memcheck tool could have reported memory leaks when TensorRT built the engine on pre-Blackwell GPUs, especially if the model contained convolution layers.

Known Issues

Functional

• There is a known accuracy issue on the Conv layers in the SDXL network on NVIDIA B200.
• When running the FLUX Transformer model in 2048x2048 spatial dimensions, it may produce NaN outputs. This can be worked around with different spatial dimensions.
• Support for B100 and B200 on Windows is considered experimental. Some networks may fail to run due to missing kernels for this GPU and OS combination. We plan to improve this support in a future release, but its status will remain experimental at this time.
• When running OSS demoBERT FP16 inference on H20 GPUs, different batch sizes may generate different outputs given the same input values. This can be worked around by using a fixed batch size.
• There is a known accuracy issue running certain networks on NVIDIA HGX H20.
• Inputs to the IRecurrenceLayer must always have the same shape. This means that ONNX models with loops whose recurrence inputs change shapes will be rejected.
• CUDA compute sanitizer may report racecheck hazards for some legacy kernels. However, related kernels do not have functional issues at runtime.
• The compute sanitizer initcheck tool may flag false positive Uninitialized __global__ memory read errors when running TensorRT applications on NVIDIA Hopper GPUs. These errors can be safely ignored and will be fixed in an upcoming CUDA release.
• Multihead attention fusion might not happen and affect performance if the number of heads is small.
• An occurrence of use-after-free in NVRTC has been fixed in CUDA 12.1. When using NVRTC from CUDA 12.0 together with the TensorRT static library, you may encounter a crash in certain scenarios. Linking the NVRTC and PTXJIT compiler from CUDA 12.1 or newer will resolve this issue.
• There are known issues reported by the Valgrind memory leak check tool when detecting potential memory leaks from TensorRT applications. The recommendation to suppress the issues is to provide a Valgrind suppression file with the following contents when running the Valgrind memory leak check tool. Add the option --keep-debuginfo=yes to the Valgrind command line to suppress these errors.
  {
  Memory leak errors with dlopen.
  Memcheck:Leak
  match-leak-kinds: definite
  ...
  fun:dlopen
  ...
  }
  {
  Memory leak errors with nvrtc
  Memcheck:Leak
  match-leak-kinds: definite
  fun:malloc
  obj:libnvrtc.so
  ...
  }
• SM 7.5 and earlier devices may not have INT8 implementations for all layers with Q/DQ nodes. In this case, you will encounter a could not find any implementation error while building your engine. To resolve this, remove the Q/DQ nodes, which quantize the failing layers.
• Installing the cuda-compat-11-4 package may interfere with CUDA-enhanced compatibility and cause TensorRT to fail even when the driver is r465. The workaround is to remove the cuda-compat-11-4 package or upgrade the driver to r470.
• For some networks, using a batch size of 4096 may cause accuracy degradation on DLA.
• For broadcasting elementwise layers running on DLA with GPU fallback enabled with one NxCxHxW input and one Nx1x1x1 input, there is a known accuracy issue if at least one of the inputs is consumed in kDLA_LINEAR format. It is recommended to explicitly set the input formats of such elementwise layers to different tensor formats.
• Exclusive padding with kAVERAGE pooling is not supported.
• Asynchronous CUDA calls are not supported in the user-defined processDebugTensor function for the debug tensor feature due to a bug in Windows 10.
• inplace_add mini-sample of the quickly_deployable_plugins Python sample may produce incorrect outputs on Windows. This will be fixed in a future release.
• When linking with libcudart_static.a using a RedHat gcc-toolset-11 or earlier compiler, you may encounter an issue where exception handling isn’t working. When a throw or exception happens, the catch is ignored, and an abort is raised, killing the program. This may be related to a linker bug causing the eh_frame_hdr ELF segment to be empty. You can workaround this issue using a new linker, such as the one from gcc-toolset-13.
• TensorRT may exit if inputs with invalid values are provided to the RoiAlign plugin (ROIAlign_TRT), especially if there is inconsistency in the indices specified in the batch_indices input and the actual batch size used.
• In the Validate against Ground Truth section of the efficientnet samples, the link to download Caffe’s ILSVRC2012 auxiliary package is unstable. Therefore, the download might fail intermittently.
• The ONNX specification of the NonMaxSuppression operation requires the iou_threshold parameter to be in the range of [0.0-1.0]. However, TensorRT does not validate the value of the parameter; therefore, TensorRT will accept values outside of this range, in which case, the engine will continue executing as if the value was capped at either end of this range.

Performance

• There is an accuracy issue when running demo/Diffusion SDXL+ControlNet on B200 GPUs.
• Up to 40% ExecutionContext memory regression compared to TensorRT 10.9 for some networks with FP16 precisions in NVIDIA Ada and Hopper GPUs.
• Up to 20% performance regression compared to TensorRT 10.8 for networks with concatenation nodes that have 100+ inputs. This will be fixed in TensorRT 10.11.
• Up to 16% performance regression compared to TensorRT 10.9 for networks with Conv+LeakyReLU, Conv+Swith, and Conv+GeLU in TF32 and FP16 precisions on SM120 Blackwell GPUs. This will be fixed in TensorRT 10.11.
• Up to 6% performance regression compared to TensorRT 10.9 for ConvNext in INT8 precision on Hopper and Ampere GPUs.
• Up to 26% performance regression for a particular version of GPT-2 which has a large concatenation at the end of the network.
• CPU peak memory usage regression with the roberta_base engine on Ampere GPUs compared to TensorRT 10.7.
• Up to 10% performance regression for Megatron networks in FP32 precision compared to TensorRT 10.8 for BS4.
• Up to 100 MB context memory size regression compared to TensorRT 8.6 on Hopper GPUs for CRNN (Convolutional Recurrent Neural Network) models. Inference performance is not affected.
• Up to 9% inference performance regression for StableDiffusion v2.0/2.1 VAE network in FP16 precision on Hopper GPUs compared to TensorRT 10.6 in CUDA 11.8 environment. This issue can be fixed by upgrading CUDA to 12.6.
• Up to 60% performance regression compared to TensorRT 8.6 on Ampere GPUs for group convolutions with N channels per group, where N is not a power of 2. This can be worked around by padding N to the next power of 2
• Up to 22% context memory size regression for HiFi-GAN networks in INT8 precision compared to TensorRT 10.5 on Ampere GPUs.
• Up to 7% performance regression for Megatron networks in FP16 precision compared to TensorRT 10.6 for BS1 and Seq128 on H100 GPUs.
• Up to 10% performance regression for BERT networks exported from TensorFlow2 in FP16 precision compared to TensorRT 10.4 for BS1 and Seq128 on A16 GPUs.
• Up to 16% regression in context memory usage for StableDiffusion XL VAE network in FP8 precision on H100 GPUs compared to TensorRT 10.3 due to a necessary functional fix.
• Up to 15% regressing in context memory usage for networks containing InstanceNorm and Activation ops compared to TensorRT 10.0.
• Up to 15% CPU memory usage regression for mbart-cnn/mamba-370m in FP16 precision and OOTB mode on NVIDIA Ada Lovelace GPUs compared to TensorRT 10.2.
• Up to 6% performance regression for BERT/Megatron networks in FP16 precision compared to TensorRT 10.2 for BS1 and Seq128 on H100 GPUs.
• Up to 6% performance regression for Bidirectional LSTM in FP16 precision on H100 GPUs compared to TensorRT 10.2.
• Performance gaps between engines built with REFIT enabled and engines built with REFIT disabled.
• Up to 60 MB engine size fluctuations for the BERT-Large INT8-QDQ model on Orin due to unstable tactic selection among tactics.
• Up to 16% performance regression for BasicUNet, DynUNet, and HighResNet in INT8 precision compared to TensorRT 9.3.
• Up to 40-second increase in engine building for BART networks on NVIDIA Hopper GPUs.
• Up to 20-second increase in engine building for some large language models (LLMs) on NVIDIA Ampere GPUs.
• Up to 2.5x build time increase compared to TensorRT 9.0 for certain Bert-like models due to additional tactics available for evaluation.
• Up to 13% performance drop for the CortanaASR model on NVIDIA Ampere GPUs compared to TensorRT 8.5.
• Up to 18% performance drop for the ShuffleNet model on A30/A40 compared to TensorRT 8.5.1.
• Convolution on a tensor with an implicitly data-dependent shape may run slower than on other tensors of the same size. Refer to the Glossary for the definition of implicitly data-dependent shapes.
• Up to 5% performance drop for networks using sparsity in FP16 precision.
• Up to 6% performance regression compared to TensorRT 8.5 on OpenRoadNet in FP16 precision on NVIDIA A10 GPUs.
• Up to 70% performance regression compared to TensorRT 8.6 on BERT networks in INT8 precision with FP16 disabled on L4 GPUs. Enable FP16 and disable INT8 in the builder config to work around this.
• In explicitly quantized networks, a group convolution with a Q/DQ pair before but no Q/DQ pair after runs with INT8-IN-FP32-OUT mixed precision. However, NVIDIA Hopper may fall back to FP32-IN-FP32-OUT if the input channel count is small.
• The kREFIT and kREFIT_IDENTICAL have performance regressions compared with non-refit engines where convolution layers are present within a branch or loop, and the precision is FP16/INT8. This issue will be addressed in future releases.