Track input and output token count for every request

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

8.16k stars 903 forks source link

Track input and output token count for every request #1718

Open brarj413 opened 3 months ago

brarj413 commented 3 months ago

hello, Please can anyone help me about tracking of input and output token counts for every request ? I am not able to add new response variable?

hijkzzz commented 2 months ago

Investigating...

sugam-nexusflow commented 4 days ago

Any update on this?