1 Pull request merged by 1 person

• 10.2 GA release update

  #3998 merged Jul 11, 2024

1 Pull request opened by 1 person

• Added support for NVIDIA GeForce RTX 4090

  #4000 opened Jul 11, 2024

5 Issues closed by 5 people

• x

  #4003 closed Jul 11, 2024

• Error[1]: [runner.cpp::execute::718] Error Code 1: Myelin (Final synchronize failed (700))

  #3995 closed Jul 10, 2024

• ICudaEngine API error

  #3973 closed Jul 10, 2024

• [defaultAllocator.cpp::deallocateAsync::64] Error Code 1: Cuda Runtime (operation not supported)

  #3980 closed Jul 9, 2024

• trt.Builder failure of TensorRT 8.5.1.7 when running python3.9.18 and python3.8.18 (miniconda) on GPU Tesla T4

  #3988 closed Jul 9, 2024

17 Issues opened by 17 people

• Inconsistencies in output when running inference on DeBERTa with disentangled attention plugin and dynamic sequence length for input

  #4005 opened Jul 11, 2024

• [Issue still exists in TRT 10.1] Fail to build the engine if input tensor number > 2^31

  #4004 opened Jul 11, 2024

• PTQ support for ViT models

  #4002 opened Jul 11, 2024

• Does TensorRT support lora and controlnet togehter?

  #4001 opened Jul 11, 2024

• wrong results of TensorRT 10.0 when running on GPU Tesla T4

  #3999 opened Jul 11, 2024

• AttributeError: 'NoneType' object has no attribute 'is_sdxl'

  #3997 opened Jul 10, 2024

• Colcon build failure of TensorRT using python3.8 and tensorrt-10.1.0.27

  #3996 opened Jul 10, 2024

• Recommended Torch Quantization Library to Use -- Modelopt v.s. Pytorch-quantization

  #3994 opened Jul 9, 2024

• TensorRT8.6.1.6 Inference cost too much time

  #3993 opened Jul 9, 2024

• Error[3]: [executionContext.cpp::nvinfer1::rt::ExecutionContext::enqueueV3::2666] Error Code 3: API Usage Error (Parameter check failed at: executionContext.cpp::nvinfer1::rt::ExecutionContext::enqueueV3::2666, condition: mContext.profileObliviousBindings.at(profileObliviousIndex) || getPtrOrNull(mOutputAllocators, profileObliviousIndex)

  #3992 opened Jul 9, 2024

• Do get_binding_shape(binding), get_binding_dtype(binding) and binding_is_input(binding) have an equivalent in TensorRT 10.2?

  #3991 opened Jul 9, 2024

• XXX failure of TensorRT X.Y when running XXX on GPU XXX

  #3990 opened Jul 8, 2024

• pycuda._driver.LogicError: cuMemHostAlloc failed: OS call failed or operation not supported on this OS

  #3989 opened Jul 8, 2024

• How to enable FP8 convolution in TensorRT 10.2

  #3987 opened Jul 7, 2024

• [ONNX Graph Surgeon] Sequence/map types are not handled

  #3986 opened Jul 6, 2024

• UINT8-to-FLOAT cast after transpose breaks the graph.

  #3985 opened Jul 5, 2024

• SDXL int8 Failure of TensorRT 10 when running txt2img_xl.py on GPU A100

  #3984 opened Jul 5, 2024

3 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• How can I use the trex tool to analyze the engine generated by trtllm

  #3982 commented on Jul 9, 2024 • 0 new comments

• TensorRT for SDXL demo not work well than use diffusers only ，consumes a lot of memory

  #3981 commented on Jul 9, 2024 • 0 new comments

• Engine building failure of TensorRT 10.2.0 (pip install) when building a custom diffusion model on RTX 4090

  #3983 commented on Jul 9, 2024 • 0 new comments