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
8.67k stars 990 forks source link

Will KOSMOS-2.5 be supported? #2106

Closed sky-fly97 closed 2 days ago

sky-fly97 commented 3 months ago

KOSMOS-2.5 is now supported in transformers, will TensorRT-LLM support it?

AdamzNV commented 2 months ago

Sorry, it's not in our consideration so far. Downloads last month of KOSMOS-2.5 looks so low. https://huggingface.co/microsoft/kosmos-2.5