Repository navigation
Releases
Releases · NVIDIA/TensorRT
TensorRT 11.5 Release
For more information, see the 11.5 release notes
-
General
- Updated default CUDA version to 13.5.
-
Parsers
- Added support for ONNX opset 25 and updated the supported ONNX release to 1.22.0.
- Updated the CMake build to locate CUDA with
find_package(CUDAToolkit); use-DCUDAToolkit_ROOTto point at a non-default CUDA install instead of-DCUDA_TOOLKIT_ROOT_DIR. - Added support for Squeeze when
axisinput is omitted.
-
Plugins
- Removed the C++17 fallback for the
PLUGIN_VALIDATEmacros - plugin sources now require C++20.
- Removed the C++17 fallback for the
-
Samples
- Added
--independentWeightLayoutsoption totrtexec. - Added
--saveEngineSoand--loadEngineSooptions totrtexecfor safety engines (requires--safe). - Updated the safety samples to use the CPU-only builder flag.
- Added
-
Tools
- Bumped Polygraphy version to v0.53.9.
- Added
model_doc_stringto onnx-graphsurgeonGraphto forward the ONNXModelProtodoc_stringon export.
TensorRT 11.4 Release
For more information, see the 11.4 release notes:
- Parsers
- Added IRefitterObserver class and IParser::setRefitObserver to better handle refittable weights when parsing.
- Plugins
- Added various C++20 updates to plugin source code.
- Samples
- Moved samples/common files only relevant to trtexec to samples/trtexecCommon.
TensorRT 11.3 Release
For more information, see the 11.3 release notes:
General
- Updated default CUDA version to 13.4
Demos
- Removed demoDiffusion.
Parsers
- Refactored how external weights are handled in IParserRefitter
Samples
- Removed the detectron2 python sample
Tools
- Bumped Polygraphy version to v0.53.6
TensorRT 11.2 Release
For more information, see the TensorRT 11.2 Release Notes.
Samples
- Added a new python sample sample_plugin_v2_to_v3_migration to showcase how to migrate from IPluginV2 to IPluginV3.
Plugins
- Added a new FFTPlugin, a cuFFT-backed plugin for complex-to-complex, real-to-complex, and complex-to-real transforms, to support the ONNX DFT operator.
Parsers
- Added IRefitterObserver class for better refitting of ONNX models.
- Added support for the DFT operator and 5D GridSample operators.
TensorRT 11.1 Release
For more information, see the TensorRT 11.1 Release Notes
General
- Default CUDA version updated to 13.3.
- Added Ubuntu 26.04 container.
- Added support for Python 3.14.
- Added agent skills for various TensorRT workflows.
Samples
- Added new sample cute_dsl_plugin.
- Added a global performance tuner option in trtexec.
Plugins
- Added new plugin topkLastDimPlugin.
Parsers
- Updated TRT_Attention to support raggedness.
TensorRT 11.0 Release
For more information, see the TensorRT 11.0 Release Notes
General
- As a new major version bump, TensorRT 11.0 brings many enhancements for its users. For full information, see the release notes.
Demos
- The BERT demo has been removed.
Samples
- Migrated
samples/python/python_pluginandsamples/python/onnx_custom_plugin/pluginto use V3 plugins
Plugins
- Added a new Faster Air Top-K plugin
- Removed deprecated plugins:
batchTile,clip,coordConvAC,cropAndResize,gelu,leakyRelu,normalize,singleStepLSTM,specialSlice,split,nms, andproposal
Parsers
- Added support for
Swishoperator - Added support for custom operators
TRT_Attention,TRT_MoE, andTRT_KVCacheUpdate
TensorRT 10.16 Release
For more information, see the TensorRT 10.16 Release Notes
General
- Default CUDA version updated to CUDA 13.2.
Samples
- Added sampleDistCollective sample to showcase multi-device execution in TensorRT.
Parsers
- Added kADJUST_FOR_DLA flag to adjust parsing behavior for ONNX models to be more amenable for DLA hardware execution.
- Added DistCollective operator support for multi-device execution in TensorRT.
TensorRT 10.15 Release
For more information, see the TensorRT 10.15 Release Notes:
Sample changes
- Added 2 safety samples sampleSafeMNIST, and sampleSafePluginV3 to demonstrate how to use TensorRT with the safety workflow.
- Added trtSafeExec to accompany the safety workflow release.
- Added python/stream_writer to showcase how to serialize a TensorRT engine directly to a custom stream using the IStreamWriter interface, rather than writing to a file or a contiguous memory buffer.
- Added python/strongly_type_autocast to demonstrate how to convert FP32 ONNX models to mixed precision (FP32-FP16) using ModelOpt's AutoCast tool and subsequently building the engine with TensorRT's Strong Typing mode.
- Added sampleCudla to demonstrate how to use the cuDLA API to run TensorRT engines on the Deep Learning Accelerator (DLA) hardware, which is available on NVIDIA Jetson and DRIVE platforms.
- Deprecated sampleCharRNN.
Plugin changes
- Deprecated bertQKVToContextPlugin and will be removed in a future release. No alternatives are planned to be provided.
Parser changes
- Added support for RotaryEmbedding, RMSNormalization and TensorScatter for improved LLM model support
- Added more specialized quantization ops for models quantized through TensorRT ModelOptimizer.
- Added kREPORT_CAPABILITY_DLA flag to enable per-node validation when building DLA engines through TensorRT.
- Added kENABLE_PLUGIN_OVERRIDE flag to enable TensorRT plugin override for nodes that share names with user plugins.
- Improved error reporting for models with multiple subgraphs, such as Loop or Scan nodes.
Demo changes
- demoDiffusion: Stable Diffusion 1.5, 2.0 and 2.1 pipelines have been deprecated and removed.
TensorRT 10.14 Release
10.14 GA - 2025-11-7
-
Sample changes
- Replace all pycuda usages with cuda-python APIs
- Removed the efficientnet samples
- Deprecated tensorflow_object_detection and efficientdet samples
- Samples will no longer be released with the packages. The TensorRT GitHub repository will be the single source.
-
Parsers:
- Added support for the
Attentionoperator - Improved refit for
ConstantOfShapenodes
- Added support for the
-
Demos
- demoDiffusion:
- Added support for the Cosmos-Predict2 text2image and video2world pipelines
- demoDiffusion:
TensorRT 10.13.3 Release
See the TensorRT 10.13.3 Release Notes for more information.
- Added support for TensorRT API Capture and Replay feature, see the developer guide for more information.
Demo changes
- Added support for Flux Kontext pipeline.