LLM 速度测试

本文介绍如何使用速度测试工具测试大语言模型的速度。

#安装 Docker

bash

curl -fsSL https://get.docker.com -o get-docker.sh && sudo sh get-docker.sh

#在 reComputer RK3576 上运行

bash

sudo docker run -it  \
        --privileged \
        --net=host \
        --device /dev/dri \
        --device /dev/dma_heap \
        --device /dev/rknpu \  
        --device /dev/mali0 \
        -v /dev:/dev \
        ghcr.io/seeed-projects/rk3576-deepseek-r1-distill-qwen:1.5b-fp16-latest

#在 reComputer RK3588 上运行

bash

sudo docker run -it  \
        --privileged \
        --net=host \
        --device /dev/dri \
        --device /dev/dma_heap \
        --device /dev/rknpu \  
        --device /dev/mali0 \
        -v /dev:/dev \
        ghcr.io/seeed-projects/rk3588-deepseek-r1-distill-qwen:1.5b-fp16-latest

#安装工具

bash

wget https://github.com/Seeed-Projects/reComputer-RK-LLM/raw/refs/heads/main/tools/llm_speed_test.py

#创建环境

bash

python -m venv .env && source .enb/bin/activate
pip install -r requests

#测试速度

bash

python llm_speed_test.py