NVIDIA TensorRT Model Optimizer

A Library to Quantize and Compress Deep Learning Models for Optimized Inference on GPUs

Examples | Benchmark Results | Documentation

Latest News

• [2024/06/03] Model Optimizer now has an experimental feature to deploy to vLLM as part of our effort to support popular deployment frameworks. Check out the workflow here
• [2024/05/08] Announcement: Model Optimizer Now Formally Available to Further Accelerate GenAI Inference Performance
• [2024/03/27] Model Optimizer supercharges TensorRT-LLM to set MLPerf LLM inference records
• [2024/03/18] GTC Session: Optimize Generative AI Inference with Quantization in TensorRT-LLM and TensorRT
• [2024/03/07] Model Optimizer's 8-bit Post-Training Quantization enables TensorRT to accelerate Stable Diffusion to nearly 2x faster
• [2024/02/01] Speed up inference with Model Optimizer quantization techniques in TRT-LLM

Table of Contents

• Model Optimizer Overview
• Installation
• Techniques
  • Quantization
  • Sparsity
• Examples
• Support Matrix
• Benchmark
• Release Notes

Model Optimizer Overview

Minimizing inference costs presents a significant challenge as generative AI models continue to grow in complexity and size. The NVIDIA TensorRT Model Optimizer (referred to as Model Optimizer, or ModelOpt) is a library comprising state-of-the-art model optimization techniques including quantization and sparsity to compress models. It accepts a torch or ONNX model as inputs and provides Python APIs for users to easily stack different model optimization techniques to produce an optimized quantized checkpoint. Seamlessly integrated within the NVIDIA AI software ecosystem, the quantized checkpoint generated from Model Optimizer is ready for deployment in downstream inference frameworks like TensorRT-LLM or TensorRT. Further integrations are planned for NVIDIA NeMo and Megatron-LM for training-in-the-loop optimization techniques. For enterprise users, the 8-bit quantization with Stable Diffusion is also available on NVIDIA NIM.

Model Optimizer is available for free for all developers on NVIDIA PyPI. This repository is for sharing examples and GPU-optimized recipes as well as collecting feedback from the community.

Installation

PIP

pip install "nvidia-modelopt[all]~=0.15.0" --extra-index-url https://pypi.nvidia.com

See the installation guide for more fine-grained control over the installation.

Docker

After installing the NVIDIA Container Toolkit, please run the following commands to build the Model Optimizer example docker container.

# Build the docker
docker/build.sh

# Obtain and start the basic docker image environment.
# The default built docker image is docker.io/library/modelopt_examples:latest
docker run --gpus all -it --shm-size 20g --rm docker.io/library/modelopt_examples:latest bash

# Check installation
python -c "import modelopt"

Alternatively for PyTorch, you can also use NVIDIA NGC PyTorch container with Model Optimizer pre-installed starting from 24.06 container. Make sure to update the Model Optimizer version to the latest one if not already.

Techniques

Quantization

Quantization is an effective model optimization technique for large models. Quantization with Model Optimizer can compress model size by 2x-4x, speeding up inference while preserving model quality. Model Optimizer enables highly performant quantization formats including FP8, INT8, INT4, etc and supports advanced algorithms such as SmoothQuant, AWQ, and Double Quantization with easy-to-use Python APIs. Both Post-training quantization (PTQ) and Quantization-aware training (QAT) are supported.

Sparsity

Sparsity is a technique to further reduce the memory footprint of deep learning models and accelerate the inference. Model Optimizer provides Python API mts.sparsity() to apply weight sparsity to a given model. mts.sparsity() supports NVIDIA 2:4 sparsity pattern and various sparsification methods, such as NVIDIA ASP and SparseGPT.

Examples

• PTQ for LLMs covers how to use Post-training quantization (PTQ) and export to TensorRT-LLM for deployment for popular pre-trained models from frameworks like
  • Hugging Face
  • NVIDIA NeMo
  • NVIDIA Megatron-LM
  • Medusa
• PTQ for Diffusers walks through how to quantize a diffusion model with FP8 or INT8, export to ONNX, and deploy with TensorRT. The Diffusers example in this repo is complementary to the demoDiffusion example in TensorRT repo and includes FP8 plugins as well as the latest updates on INT8 quantization.
• QAT for LLMs demonstrates the recipe and workflow for Quantization-aware Training (QAT), which can further preserve model accuracy at low precisions (e.g., INT4, or 4-bit in NVIDIA Blackwell platform).
• Sparsity for LLMs shows how to perform Post-training Sparsification and Sparsity-aware fine-tuning on a pre-trained Hugging Face model.
• ONNX PTQ shows how to quantize the ONNX models in INT4 or INT8 quantization mode. The examples also include the deployment of quantized ONNX models using TensorRT.

Support Matrix

• For LLMs, please refer to this support matrix.
• For Diffusion, the Model Optimizer supports Stable Diffusion 1.5, Stable Diffusion XL, and SDXL-Turbo.

Benchmark

Please find the benchmarks here.

Release Notes

Please see Model Optimizer Changelog here.