Here are the classes, structs, unions and interfaces with brief descriptions:

[detail level 123]

▼Nnvinfer1	The TensorRT API version 1 namespace
▼Nimpl
CEnumMaxImpl	Declaration of EnumMaxImpl struct to store the exclusive upper bound of an enumeration type
CEnumMaxImpl< ActivationType >
CEnumMaxImpl< AllocatorFlag >	Maximum number of elements in AllocatorFlag enum
CEnumMaxImpl< APILanguage >	One greater than the maximum value of APILanguage enum
CEnumMaxImpl< AttentionIOForm >
CEnumMaxImpl< AttentionNormalizationOp >
CEnumMaxImpl< BoundingBoxFormat >
CEnumMaxImpl< BuilderFlag >
CEnumMaxImpl< CausalMaskKind >
CEnumMaxImpl< CollectiveOperation >
CEnumMaxImpl< CudaGraphStrategy >	Maximum number of CUDA graph strategies in CudaGraphStrategy enum
CEnumMaxImpl< CumulativeOperation >	Maximum number of elements in CumulativeOperation enum
CEnumMaxImpl< DataType >	One greater than the maximum value of DataType enum
CEnumMaxImpl< DeviceType >	Maximum number of elements in DeviceType enum
CEnumMaxImpl< DimensionOperation >	Maximum number of elements in DimensionOperation enum
CEnumMaxImpl< DynamicShapesKernelSpecializationStrategy >
CEnumMaxImpl< ElementWiseOperation >
CEnumMaxImpl< EngineCapability >	Maximum number of elements in EngineCapability enum
CEnumMaxImpl< EngineStat >	Maximum number of engine statistic kinds in EngineStat enum
CEnumMaxImpl< EngineValidity >	Maximum number of elements in EngineValidity enum
CEnumMaxImpl< ErrorCode >	One greater than the maximum value of ErrorCode enum
CEnumMaxImpl< ExecutionContextAllocationStrategy >
CEnumMaxImpl< FillOperation >
CEnumMaxImpl< GatherMode >
CEnumMaxImpl< HardwareCompatibilityLevel >
CEnumMaxImpl< ILogger::Severity >	Maximum number of elements in ILogger::Severity enum
CEnumMaxImpl< InterpolationMode >
CEnumMaxImpl< KVCacheMode >
CEnumMaxImpl< LayerInformationFormat >	Maximum number of layer information formats in LayerInformationFormat enum
CEnumMaxImpl< LayerType >
CEnumMaxImpl< LoopOutput >
CEnumMaxImpl< MatrixOperation >
CEnumMaxImpl< MemoryPoolType >
CEnumMaxImpl< MoEActType >
CEnumMaxImpl< NetworkDefinitionCreationFlag >
CEnumMaxImpl< OptProfileSelector >	Number of different values of OptProfileSelector enum
CEnumMaxImpl< PaddingMode >
CEnumMaxImpl< PoolingType >
CEnumMaxImpl< PreviewFeature >
CEnumMaxImpl< ProfilingVerbosity >	Maximum number of profile verbosity levels in ProfilingVerbosity enum
CEnumMaxImpl< ReduceOperation >
CEnumMaxImpl< ResizeCoordinateTransformation >
CEnumMaxImpl< ResizeRoundMode >
CEnumMaxImpl< ResizeSelector >
CEnumMaxImpl< RuntimePlatform >
CEnumMaxImpl< SampleMode >
CEnumMaxImpl< ScaleMode >
CEnumMaxImpl< ScatterMode >
CEnumMaxImpl< SerializationFlag >	Maximum number of serialization flags in SerializationFlag enum
CEnumMaxImpl< TacticSource >	Maximum number of tactic sources in TacticSource enum
CEnumMaxImpl< TempfileControlFlag >	Maximum number of elements in TempfileControlFlag enum
CEnumMaxImpl< TensorFormat >	Maximum number of elements in TensorFormat enum
CEnumMaxImpl< TensorIOMode >	One greater than the maximum value of TensorIOMode enum
CEnumMaxImpl< TensorLocation >	Maximum number of elements in TensorLocation enum
CEnumMaxImpl< TilingOptimizationLevel >
CEnumMaxImpl< TopKOperation >
CEnumMaxImpl< TripLimit >
CEnumMaxImpl< UnaryOperation >
CEnumMaxImpl< WeightsRole >	Maximum number of elements in WeightsRole enum
▼Nplugin
CGridAnchorParameters	The Anchor Generator plugin layer generates the prior boxes of designated sizes and aspect ratios across all dimensions (H x W). GridAnchorParameters defines a set of parameters for creating the plugin layer for all feature maps
CPriorBoxParameters	The PriorBox plugin layer generates the prior boxes of designated sizes and aspect ratios across all dimensions (H x W)
CRegionParameters	The Region plugin layer performs region proposal calculation
CRPROIParams	RPROIParams is used to create the RPROIPlugin instance
CsoftmaxTree	When performing yolo9000, softmaxTree is helping to do softmax on confidence scores, for element to get the precise classification through word-tree structured classification definition
▼Nv_1_0	Forward declare IErrorRecorder and ILogger for use in other interfaces
CIDebugListener
CIErrorRecorder
CIGpuAllocator
CIGpuAsyncAllocator
CILogger	Application-implemented logging interface for the builder, refitter and runtime
CILoggerFinder	A virtual base class to find a logger. Allows a plugin to find an instance of a logger if it needs to emit a log message. A pointer to an instance of this class is passed to a plugin shared library on initialization when that plugin is serialized as part of a version-compatible plan. See the plugin chapter in the developer guide for details
CIOutputAllocator
CIPluginCapability
CIPluginCreator
CIPluginCreatorInterface
CIPluginCreatorV3One
CIPluginResource
CIPluginV3
CIPluginV3OneBuild
CIPluginV3OneCore
CIPluginV3OneRuntime
CIProfiler
CIProgressMonitor
CIStreamReader
CIStreamReaderV2
CIStreamWriter
CTimingCacheKey	The key to retrieve timing cache entries
CTimingCacheValue
▼Nv_2_0
CIPluginV3OneBuild
CDims2	Descriptor for two-dimensional data
CDims3	Descriptor for three-dimensional data
CDims4	Descriptor for four-dimensional data
CDims64
CDimsExprs	Analog of class Dims with expressions instead of constants for the dimensions
CDimsHW	Descriptor for two-dimensional spatial data
CDynamicPluginTensorDesc	Summarizes tensors that a plugin might see for an input or output
CIActivationLayer	An Activation layer in a network definition
CIAssertionLayer	An assertion layer in a network
CIAttention	Helper for constructing an attention that consumes query, key and value tensors
CIAttentionBoundaryLayer	This is a base class for Attention boundary layers
CIAttentionInputLayer	This layer represents an input to an attention subgraph
CIAttentionOutputLayer	This layer represents an output of an IAttention
CIBuilder	Builds an engine from a network definition
CIBuilderConfig	Holds properties for configuring a builder to produce an engine
CICastLayer	A cast layer in a network
CIConcatenationLayer	A concatenation layer in a network definition
CIConditionLayer	This layer represents a condition input to an IIfConditional
CIConstantLayer	Layer that represents a constant value
CIConvolutionLayer	A convolution layer in a network definition
CICudaEngine	An engine for executing inference on a built network, with functionally unsafe features
CICumulativeLayer	Layer that represents a cumulative operation across a tensor
CIDeconvolutionLayer	A deconvolution layer in a network definition
CIDequantizeLayer	A Dequantize layer in a network definition
CIDimensionExpr	An IDimensionExpr represents an integer expression constructed from constants, input dimensions, and binary operations. These expressions are can be used in overrides of IPluginV2DynamicExt::getOutputDimensions or IPluginV3OneBuild::getOutputShapes() to define output dimensions in terms of input dimensions
CIDistCollectiveLayer
CIDynamicQuantizeLayer	A network layer to perform dynamic quantization
CIEinsumLayer	An Einsum layer in a network
CIElementWiseLayer	A elementwise layer in a network definition
CIEngineInspector	An engine inspector which prints out the layer information of an engine or an execution context
CIExecutionContext	Context for executing inference using an engine, with functionally unsafe features
CIExprBuilder	Object for constructing IDimensionExpr
CIFillLayer	Generate a tensor according to a specified mode
CIGatherLayer	A Gather layer in a network definition. Supports several kinds of gathering
CIGridSampleLayer	A GridSample layer in a network definition
CIHostMemory	Class to handle library allocated memory that is accessible to the user
CIIdentityLayer	A layer that represents the identity function
CIIfConditional	Helper for constructing conditionally-executed subgraphs
CIIfConditionalBoundaryLayer	This is a base class for Conditional boundary layers
CIIfConditionalInputLayer	This layer represents an input to an IIfConditional
CIIfConditionalOutputLayer	This layer represents an output of an IIfConditional
CIIteratorLayer	A layer to do iterations
CIKVCacheUpdateLayer	Layer that represents a KVCacheUpdate operation
CILayer	Base class for all layer classes in a network definition
CILoop	Helper for creating a recurrent subgraph
CILoopBoundaryLayer	This is a base class for Loop boundary layers
CILoopOutputLayer	An ILoopOutputLayer is the sole way to get output from a loop
CILRNLayer	A LRN layer in a network definition
CIMatrixMultiplyLayer	Layer that represents a Matrix Multiplication
CIMoELayer	A MoE layer in a network definition. Mixture of Experts (MoE) is a collection of experts with each expert specializing in processing different subsets of input data. The key innovation lies in using a Router that selectively activates only the specific experts needed for a given input, rather than engaging the entire neural network for every task
CINetworkDefinition	A network definition for input to the builder
CINMSLayer	A non-maximum suppression layer in a network definition
CINoCopy	Forward declaration of IEngineInspector for use by other interfaces
CINonZeroLayer
CINormalizationLayer	A normalization layer in a network definition
CInterfaceInfo	Version information associated with a TRT interface
CIOneHotLayer	A OneHot layer in a network definition
CIOptimizationProfile	Optimization profile for dynamic input dimensions and shape tensors
CIPaddingLayer	Layer that represents a padding operation
CIParametricReLULayer	Layer that represents a parametric ReLU operation
CIPluginRegistry	Single registration point for all plugins in an application. It is used to find plugin implementations during engine deserialization. Internally, the plugin registry is considered to be a singleton so all plugins in an application are part of the same global registry. Note that the plugin registry is only supported for plugins of type IPluginV2 and should also have a corresponding IPluginCreator implementation
CIPluginResourceContext	Interface for plugins to access per context resources provided by TensorRT
CIPluginV2	Plugin class for user-implemented layers
CIPluginV2DynamicExt	Similar to IPluginV2Ext, but with support for dynamic shapes
CIPluginV2Ext	Plugin class for user-implemented layers
CIPluginV2IOExt	Plugin class for user-implemented layers
CIPluginV2Layer	Layer type for pluginV2
CIPluginV3Layer	Layer type for V3 plugins
CIPoolingLayer	A Pooling layer in a network definition
CIQuantizeLayer	A Quantize layer in a network definition
CIRaggedSoftMaxLayer	A RaggedSoftmax layer in a network definition
CIRecurrenceLayer	A recurrence layer in a network definition
CIReduceLayer	Layer that represents a reduction across a non-bool tensor
CIRefitter	Updates weights in an engine
CIResizeLayer	A resize layer in a network definition
CIReverseSequenceLayer	A ReverseSequence layer in a network definition
CIRotaryEmbeddingLayer	Layer that implements Rotary Position Embedding (RoPE) (https://arxiv.org/abs/2104.09864)
CIRuntime	Allows a serialized functionally unsafe engine to be deserialized
CIRuntimeCache	A class for Runtime cache currently used for TensorRT JIT compilation. This cache can be serialized and deserialized to support persistence in disk files
CIRuntimeConfig	A class for runtime configuration. This class is used during execution context creation
CIScaleLayer	A Scale layer in a network definition
CIScatterLayer	A scatter layer in a network definition. Supports several kinds of scattering
CISelectLayer	Select elements from two data tensors based on a condition tensor
CISerializationConfig	Holds properties for configuring an engine to serialize the binary
CIShapeLayer	Layer type for getting shape of a tensor
CIShuffleLayer	Layer type for shuffling data
CISliceLayer	Slices an input tensor into an output tensor based on the offset and strides
CISoftMaxLayer	A Softmax layer in a network definition
CISqueezeLayer	Layer that represents a squeeze operation, removing unit dimensions of the first input tensor on a set of axes specified by the second input tensor
CITensor	A tensor in a network definition
CITimingCache	Class to handle tactic timing info collected from builder
CITopKLayer	Layer that represents a TopK reduction
CITripLimitLayer	A layer that represents a trip-count limiter
CIUnaryLayer	Layer that represents an unary operation
CIUnsqueezeLayer	Layer that represents an unsqueeze operation, which reshapes the first input tensor by inserting unit-length dimensions to the output at the axes specified by the second input tensor
CIVersionedInterface	An Interface class for version control
CPermutation	Represents a permutation of dimensions
CPluginField	Structure containing plugin attribute field names and associated data This information can be parsed to decode necessary plugin metadata
CPluginFieldCollection	Plugin field collection struct
CPluginRegistrar	Register the plugin creator to the registry The static registry object will be instantiated when the plugin library is loaded. This static object will register all creators available in the library to the registry
CPluginTensorDesc	Fields that a plugin might see for an input or output
CWeights	An array of weights used as a layer parameter
▼Nnvonnxparser	The TensorRT ONNX parser API namespace
CIOnnxConfig	Configuration Manager Class
CIParser	Object for parsing ONNX models into a TensorRT network definition
CIParserError	Object containing information about an error
CIParserRefitter	An interface designed to refit weights from an ONNX model
CDims	Structure to define the dimensions of a tensor
CIDebugListener	User-implemented callback for notification when value of a debug tensor is updated
CIErrorRecorder	Reference counted application-implemented error reporting interface for TensorRT objects
CIGpuAllocator	Application-implemented class for controlling allocation on the GPU
CIGpuAsyncAllocator	Application-implemented class for controlling asynchronous (stream ordered) memory allocation on the GPU
CINonZero	A NonZero layer in a network
CIOutputAllocator	Callback from ExecutionContext::enqueueV3()
CIPluginCapability	Base class for plugin capability interfaces
CIPluginCreator	Plugin creator class for user implemented layers
CIPluginCreatorInterface	Base class for all plugin creator versions
CIPluginCreatorV3One	A plugin creator class capable of producing IPluginV3 objects
CIPluginResource	Interface for plugins to define custom resources that could be shared through the plugin registry
CIPluginV3	Plugin class for the V3 generation of user-implemented layers
CIPluginV3OneBuild	A plugin capability interface that enables the build capability (PluginCapabilityType::kBUILD). Exposes methods that allow the expression of the build time properties and behavior of a plugin
CIPluginV3OneBuildV2	A plugin capability interface that extends IPluginV3OneBuild by providing I/O aliasing functionality
CIPluginV3OneCore	A plugin capability interface that enables the core capability (PluginCapabilityType::kCORE)
CIPluginV3OneRuntime	A plugin capability interface that enables the runtime capability (PluginCapabilityType::kRUNTIME). Exposes methods that allow the expression of the runtime properties and behavior of a plugin
CIProfiler	Application-implemented interface for profiling
CIProgressMonitor	Application-implemented progress reporting interface for TensorRT
CIStreamReader	Application-implemented class for reading data in a stream-based manner
CIStreamReaderV2	Application-implemented class for reading data in a stream-based manner asynchronously. Intended for use with the GDS API for optimizing load times
CIStreamWriter	Application-implemented class for writing data in a stream-based manner
CPluginVersion	Definition of plugin versions
CValue	The values in the cache entry