NVIDIA-NCCL下载资源分享，跳过Authenticator验证

内卷焦虑人士2024-06-25 12:28

目录

  • Authenticator
  • 进入下载页面
    • [Download NCCL 2.22.3, for CUDA 12.5, June 18th, 2024](#Download NCCL 2.22.3, for CUDA 12.5, June 18th, 2024)
      • [Local installers (x86)](#Local installers (x86))
      • [Network installers (x86)](#Network installers (x86))
      • [Local installers (ARM)](#Local installers (ARM))
      • [Network installers (ARM)](#Network installers (ARM))
    • [Download NCCL 2.22.3, for CUDA 12.4, June 18th, 2024](#Download NCCL 2.22.3, for CUDA 12.4, June 18th, 2024)
      • [Local installers (x86)](#Local installers (x86))
      • [Network installers (x86)](#Network installers (x86))
      • [Local installers (ARM)](#Local installers (ARM))
      • [Network installers (ARM)](#Network installers (ARM))
    • [Download NCCL 2.22.3, for CUDA 12.2, June 18th, 2024](#Download NCCL 2.22.3, for CUDA 12.2, June 18th, 2024)
      • [Local installers (x86)](#Local installers (x86))
      • [Network installers (x86)](#Network installers (x86))
      • [Local installers (ARM)](#Local installers (ARM))
  • 如果需要其他版本请留言

Authenticator

现在NVIDIA登录需要Authenticator验证，很多人会卡在这里导致无法下载后续的资源

如果有的话，在最下面选择已验证ID，扫描QR码，回到主页就能看到多出了一个NVIDIA的选项栏，输入验证码即可

进入下载页面

Download NCCL 2.22.3, for CUDA 12.5, June 18th, 2024

Local installers (x86)

O/S agnostic local installer
Local installer for Ubuntu 22.04
Local installer for Ubuntu 20.04
Local installer for RedHat/CentOS 9
Local installer for RedHat/CentOS 8

Network installers (x86)

If not already done, make sure you install the network repository:

bash 复制代码 
Network Installer for Ubuntu22.04

$ wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/cuda-keyring_1.0-1_all.deb
$ sudo dpkg -i cuda-keyring_1.0-1_all.deb
$ sudo apt-get update
Network Installer for Ubuntu20.04

$ wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/cuda-keyring_1.0-1_all.deb
$ sudo dpkg -i cuda-keyring_1.0-1_all.deb
$ sudo apt-get update
Network Installer for RedHat/CentOS 9

$ sudo dnf config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/rhel9/x86_64/cuda-rhel9.repo
Network Installer for RedHat/CentOS 8

$ sudo dnf config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/rhel8/x86_64/cuda-rhel8.repo

then run the following command to installer NCCL:

For Ubuntu: sudo apt install libnccl2=2.22.3-1+cuda12.5 libnccl-dev=2.22.3-1+cuda12.5

For RHEL/Centos: sudo yum install libnccl-2.22.3-1+cuda12.5 libnccl-devel-2.22.3-1+cuda12.5 libnccl-static-2.22.3-1+cuda12.5

Local installers (ARM)

O/S agnostic local installer
Local installer for Ubuntu 22.04
Local installer for Ubuntu 20.04
Local installer for RedHat/CentOS 9
Local installer for RedHat/CentOS 8

Network installers (ARM)

If not already done, make sure you install the network repository:

bash 复制代码 
Network Installer for Ubuntu22.04

$ wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/sbsa/cuda-keyring_1.0-1_all.deb
$ sudo dpkg -i cuda-keyring_1.0-1_all.deb
$ sudo apt-get update
Network Installer for Ubuntu20.04

$ wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/sbsa/cuda-keyring_1.0-1_all.deb
$ sudo dpkg -i cuda-keyring_1.0-1_all.deb
$ sudo apt-get update
Network Installer for RedHat/CentOS 9

$ sudo dnf config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/rhel9/sbsa/cuda-rhel9.repo
Network Installer for RedHat/CentOS 8

$ sudo dnf config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/rhel8/sbsa/cuda-rhel8.repo

then run the following command to installer NCCL:

For Ubuntu: sudo apt install libnccl2=2.22.3-1+cuda12.5 libnccl-dev=2.22.3-1+cuda12.5

For RHEL/Centos: sudo yum install libnccl-2.22.3-1+cuda12.5 libnccl-devel-2.22.3-1+cuda12.5 libnccl-static-2.22.3-1+cuda12.5

Download NCCL 2.22.3, for CUDA 12.4, June 18th, 2024

Local installers (x86)

O/S agnostic local installer
Local installer for Ubuntu 22.04
Local installer for Ubuntu 20.04
Local installer for RedHat/CentOS 9
Local installer for RedHat/CentOS 8

Network installers (x86)

If not already done, make sure you install the network repository:

bash 复制代码 
Network Installer for Ubuntu22.04

$ wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/cuda-keyring_1.0-1_all.deb
$ sudo dpkg -i cuda-keyring_1.0-1_all.deb
$ sudo apt-get update
Network Installer for Ubuntu20.04

$ wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/cuda-keyring_1.0-1_all.deb
$ sudo dpkg -i cuda-keyring_1.0-1_all.deb
$ sudo apt-get update
Network Installer for RedHat/CentOS 9

$ sudo dnf config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/rhel9/x86_64/cuda-rhel9.repo
Network Installer for RedHat/CentOS 8

$ sudo dnf config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/rhel8/x86_64/cuda-rhel8.repo

Local installers (ARM)

O/S agnostic local installer
Local installer for Ubuntu 22.04

Network installers (ARM)

If not already done, make sure you install the network repository:

bash 复制代码 
Network Installer for Ubuntu22.04

$ wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/sbsa/cuda-keyring_1.0-1_all.deb
$ sudo dpkg -i cuda-keyring_1.0-1_all.deb
$ sudo apt-get update

then run the following command to installer NCCL:

For Ubuntu: sudo apt install libnccl2=2.22.3-1+cuda12.4 libnccl-dev=2.22.3-1+cuda12.4

For RHEL/Centos: sudo yum install libnccl-2.22.3-1+cuda12.4 libnccl-devel-2.22.3-1+cuda12.4 libnccl-static-2.22.3-1+cuda12.4

Download NCCL 2.22.3, for CUDA 12.2, June 18th, 2024

Local installers (x86)

O/S agnostic local installer
Local installer for Ubuntu 22.04
Local installer for RedHat/CentOS 9

Network installers (x86)

If not already done, make sure you install the network repository:

bash 复制代码 
Network Installer for Ubuntu22.04

$ wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/cuda-keyring_1.0-1_all.deb
$ sudo dpkg -i cuda-keyring_1.0-1_all.deb
$ sudo apt-get update
Network Installer for RedHat/CentOS 9

$ sudo dnf config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/rhel9/x86_64/cuda-rhel9.repo

then run the following command to installer NCCL:

For Ubuntu: sudo apt install libnccl2=2.22.3-1+cuda12.2 libnccl-dev=2.22.3-1+cuda12.2

For RHEL/Centos: sudo yum install libnccl-2.22.3-1+cuda12.2 libnccl-devel-2.22.3-1+cuda12.2 libnccl-static-2.22.3-1+cuda12.2

Local installers (ARM)

O/S agnostic local installer

如果需要其他版本请留言

相关推荐
ai小陈3 天前
大模型推理显存不够怎么办？量化、KV Cache与CPU Offload优化实战
大数据·人工智能·ai·云计算·gpu算力
Lifangyun_WD4 天前
第一次租 GPU 服务器怎么开始？镜像、SSH、数据存储和停机前要做什么
服务器·云计算·gpu算力·gpu算力租赁
Akiyama_Mio-Kon6 天前
计算机每日时报（2026-08-27）：算力继续加码，Agent 先补好安全与交付边界
aws·nvidia·microsoft 365·github copilot·ai 基础设施·ai agent 安全·excel python
算力百科小智7 天前
哪些算力租用平台比较好？通过三轮试租，比查看推荐榜单更准确
gpu算力·gpu租用
ai小陈7 天前
PyTorch多GPU分布式训练实战：从单卡脚本迁移到DDP
服务器·人工智能·pytorch·分布式·深度学习·ai·gpu算力
Zhu7587 天前
GPU驱动与CUDA安装指南
nvidia·cuda
云上工程笔记8 天前
RDMA 高速互联 GPU 云怎么选：多卡训练、分布式训练与 RoCE/InfiniBand 架构对比
架构·云计算·gpu算力
gwf2169 天前
AI集群RDMA网络架构总览：从Scale-Out到多平面
rdma·nccl·scaleout·rocev2·ai集群·rnic·gpudirect
算力百科小智11 天前
H100租用平台如何选？真正拉开差距的是整机拓扑和集群网络
gpu算力·gpu租用
热门推荐
01国内可直接用、免费额度/永久免费的大模型API清单（含 SiliconFlow、火山、阿里、智谱、百度、Kimi、DeepSeek、DMXAPI 等）02GitHub 镜像站点03蜘蛛磁力 搜索引擎大全，如何使用蜘蛛磁力查找磁力链接042026 国产 AI 大模型横评：DeepSeek、通义千问、Kimi、文心一言、星火、豆包谁更能打？05历年考研数学一、数学二、数学三真题试卷及答案PDF06ZCode 周末送额度活动开启：3 亿 Token 免费领取07微信历史版本含下载地址（ Windows PC | 安卓 | MAC ）及设置微信不更新08AI 编程 IDE 全景解析 2026：Agent 全面接管开发链路09Agnes AI 免费 API 接入指南：文本、生图、生视频，一套接口全免费10VSCode+CodeX扩展，无法启动