Willow

just add some comment, please ignore it.

When I convert the swinv2 model by torch2trt, an `IndexError` raise as following: ```txt File "/usr/local/lib/python3.8/dist-packages/timm/models/swin_transformer_v2.py", line 215, in forward qkv = F.linear(input=x, weight=self.qkv.weight, bias=qkv_bias) File "/usr/local/lib/python3.8/dist-packages/torch2trt-0.4.0-py3.8.egg/torch2trt/torch2trt.py", line 310, in...

开发环境配置过程中问题频出

5

1. 这个项目有即开即用的 docker 环境 当前出现的问题有： - ModuleNotFoundError: No module named 'paddle.fluid' - ImportError: libcudart.so.10.2: cannot open shared object file: No such file or directory - 其他一些包的不兼容或者安装失败，如 ``` #7 74.65 ERROR:...

[Feature] Return most probable tokens + logprobs

2

## Motivation ### Feature request It would be great if the API could return a list of most probable tokens (along with their logprobs) for each step. This could be...

Just for better in markdown

Batchsize=40 failure of TensorRT 8.6.1 when running transformers on GPU A30

2

## Description When the input batchsize is large(such as 40, 512, 1024), the output of the model is inconsistent with onnxrt. In addition, our test found that when batchsize=32, the...

[issue] C++ runtime support multimodal model llava-one-vision

1

How to support the new model in cpp runtime ? Is there any reference document ? For example, the multimodal model [llava-one-vision](https://huggingface.co/lmms-lab/llava-onevision-qwen2-7b-ov) Following [TRTLLM doc](https://github.com/NVIDIA/TensorRT-LLM/tree/main/examples/multimodal#llava-llava-next-and-vila), some issue as following: Hardware:...