Is it "INT8 or FP8" with "--use_weight_only --weight_only_precision int8 --qformat fp8"

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.

Apache License 2.0

7.34k stars 794 forks source link

System Info

GPU - A10

Who can help?

@Tracin

Information

[X] The official example scripts
[ ] My own modified scripts

Tasks

[X] An officially supported task in the examples folder (such as GLUE/SQuAD, ...)
[ ] My own task or dataset (give details below)

Reproduction

Hi there,

With "--use_weight_only --weight_only_precision int8 --qformat fp8", will the quantization be in INT8 or FP8?

Expected behavior

FP8 dtype when printing out weights.

actual behavior

The weights are in torch.int8.

additional notes

N/A

NVIDIA / TensorRT-LLM