Awesome

LLM-Inference-Bench: Inference Benchmarking of Large Language Models on AI Accelerators

Authors: Krishna Teja Chitty-Venkata, Siddhisanket Raskar, Bharat Kale, Farah Ferdaus, Aditya Tanikanti, Ken Raffenetti, Valerie Taylor, Murali Emani, Venkatram Vishwanath

Affliation: Argonne National Laboratory

This repository is the official implementation of "LLM-Inference-Bench" paper

Table of Contents

• About
• Metrix
• Features
• Evaluated Llms
• Dashboard
• Citation
• Acknowledgement

📌 About

Large Language Models (LLMs) have propelled groundbreaking advancements across several domains and are commonly used for text generation applications. However, the computational demands of these complex models pose significant challenges, requiring efficient hardware acceleration. Benchmarking the performance of LLMs across diverse hardware platforms is crucial to understanding their scalability and throughput characteristics. We introduce LLM-Inference-Bench, a comprehensive benchmarking suite to evaluate the hardware inference performance of LLMs. We thoroughly analyze diverse hardware platforms, including GPUs from Nvidia and AMD and specialized AI accelerators, Intel Habana and SambaNova. Our evaluation includes several LLM inference frameworks and models from LLaMA, Mistral, and Qwen families with 7B and 70B parameters. Our benchmarking results reveal the strengths and limitations of various models, hardware platforms, and inference frameworks. We provide an interactive dashboard to help identify configurations for optimal performance for a given hardware platform.

Metrix of Evaluated Frameworks and Hardwares :

Evaluated LLMs

Dashboard

Please find our results dashboard to compare different LLMs, Inference Frameworks and Hardware for different batch sizes and input/output lengths here

📌 Citation

If you find this repository useful, please consider citing our paper:

@article{chitty2024llm,
  title={LLM-Inference-Bench: Inference Benchmarking of Large Language Models on AI Accelerators},
  author={Chitty-Venkata, Krishna Teja and Raskar, Siddhisanket and Kale, Bharat and Ferdaus, Farah and Tanikanti, Aditya and Raffenetti, Ken and Taylor, Valerie and Emani, Murali and Vishwanath, Venkatram},
  journal={arXiv preprint arXiv:2411.00136},
  year={2024}
}

Acknowledgements

This research used resources of the Argonne Leadership Computing Facility, a U.S. Department of Energy (DOE) Office of Science user facility at Argonne National Laboratory and is based on research supported by the U.S. DOE Office of Science-Advanced Scientific Computing Research Program, under Contract No. DE-AC02-06CH11357. We gratefully acknowledge the computing resources provided and operated by the Joint Laboratory for System Evaluation (JLSE) at Argonne National Laboratory.