NVIDIA / TensorRT-LLM

TensorRT-LLM provides users with an easy-to-use Python API to define Large Language Models (LLMs) and build TensorRT engines that contain state-of-the-art optimizations to perform inference efficiently on NVIDIA GPUs. TensorRT-LLM also contains components to create Python and C++ runtimes that execute those TensorRT engines.
https://nvidia.github.io/TensorRT-LLM
Apache License 2.0
8.71k stars 996 forks source link

Fix prompt_table_data empty tensor shape error #2470

Open BasicCoder opened 2 days ago

BasicCoder commented 2 days ago

When executing LLM in a VLM model alone, the correct prompt_table_data's hidden_size = self.hidden_size * tp_size, because self.hidden_size = pretrained_config.hidden_size // tp_size