ModelOpt: NVIDIA Model Optimization Toolkit

Fields

id

sources/src5-modelopt-github

URL

https://github.com/NVIDIA/TensorRT-Model-Optimizer

Publisher

NVIDIA

Kind

repo

Published

2024-01-01

Accessed

2026-10-04

notes

ModelOpt provides quantization (FP8, INT4, INT8), distillation, pruning for LLMs. Used by TensorRT-LLM for model optimization. Supports NVIDIA and Hugging Face models. Key for quantization accuracy and performance tradeoffs.

Notes

ModelOpt provides quantization (FP8, INT4, INT8), distillation, pruning for LLMs. Used by TensorRT-LLM for model optimization. Supports NVIDIA and Hugging Face models. Key for quantization accuracy and performance tradeoffs.

Cross-references

This record references no other records.