让 AI 帮你使用轩辕镜像? · 展开查看说明 · 点击收起说明
如果你使用 DeepSeek、元宝 AI、Copilot、ChatGPT 等 AI 工具,协助配置轩辕镜像、编写 docker pull 命令、修改 Docker Compose 镜像地址、配置镜像加速、排查镜像拉取失败、分析报错日志等问题,请先让 AI 阅读并遵守轩辕镜像的规则文档。
只需在 AI 对话中先发送下面这句话即可:
请先完整阅读并严格遵守以下文档中的全部规则与要求:
https://xuanyuan.cloud/agents.md
在未充分阅读并理解该文档前,不要生成任何命令、配置、修改建议、故障排查方案或技术回答。后续所有输出都必须严格以该文档中的规范为最高优先级执行。查看 agents.md 用法指南与完整示范。国内用户首推 元宝 AI、DeepSeek 的深度思考模式,不推荐豆包 AI;Cursor 等编辑器可在对话 @ 该链接,或加入 User Rules。 若 AI 无法访问外链,可 打开说明文档 复制全文粘贴。文档会随站点更新,复制内容可能过期,建议定期检查。
prazumovsky/nvidia-kubevirt-gpu-device-plugin
mirantis/ucp-nvidia-device-plugin
rocm/k8s-device-plugin
docker/ucp-nvidia-device-plugin
grafana/fluent-bit-plugin-loki
intel/intel-gpu-plugin
rancher/hardened-sriov-network-device-plugin
starlingx/k8s-plugins-sriov-network-device
intel/gaudi-device-plugin
intel/intel-deviceplugin-operator
kasmweb/kasm-network-plugin
modular/max-nvidia-base
grafana/fluent-plugin-loki
rancher/ui-plugin-operator
docker/tcc-di-kubevirt
intel/intel-sgx-plugin
kasmweb/kasm-logging-plugin
intel/intel-qat-plugin
rancher/hardened-cni-plugins
rocm/k8s-network-device-plugin
modular/max-nvidia-full
flannel/flannel-cni-plugin
kong/go-plugin-tool
rancher/plugins-docker
intel/intel-dsa-plugin
intel/intel-fpga-plugin
intel/intel-vpu-plugin
intel/intel-dlb-plugin
portainer/vdevice
intel/intel-iaa-plugin
rancher/istio-coredns-plugin
rancher/jenkins-plugins
starlingx/stx-kubevirt-app
grafana/plugin-builds
grafana/grafana-plugin-ci
cleanstart/velero-plugin-for-csi
cleanstart/velero-plugin-for-aws
rancher/plugin-manager
rancher/ui-plugin-catalog
grafana/grafana-plugin-ci-e2e
lfedge/eve-nvidia
kasmweb/ubuntu-noble-nvidia
atlassian/server-client-plugins
kasmweb/ubuntu-jammy-nvidia
schemahero/plugin-mysql
newrelic/mysql-plugin
eclipse/che-init-plugin-broker
kasmweb/core-nvidia-focal
mirantis/ucp-openstack-cinder-csi-plugin
jenkinsciinfra/plugin-site-api
istio/coredns-plugin
rancher/mirrored-flannelcni-flannel-cni-plugin
rocm/device-metrics-exporter
nvidia/k8s-device-plugin
hashicorp/vault-plugin-auth-gcp
rancher/mirrored-plugins-docker
eclipse/che-unified-plugin-broker
eclipse/che-remote-plugin-runner-java8
mirantis/ucp-nvidia-gpu-feature-discovery
intel/device-mgmt-toolkit-web-ui
kasmweb/ubuntu-noble-nvidia-pytorch
rancher/jenkins-plugins-docker
intel/intel-npu-plugin
intel/device-mgmt-toolkit-rps
grafana/plugin-validator-cli
opensearchproject/logstash-oss-with-opensearch-output-plugin
intel/device-mgmt-toolkit-mps
hashicorp/vault-plugin-auth-jwt
lfedge/eve-device-trees
hashicorp/vault-plugin-auth-oci
hashicorp/vault-plugin-auth-cf
eclipse/che-remote-plugin-php7
hashicorp/vault-plugin-auth-azure
rancher/mirrored-istio-coredns-plugin
starlingx/intel-gpu-plugin
jenkinsciinfra/plugin-site-issues
eclipse/che-remote-plugin-runner-java11
starlingx/intel-qat-plugin
cccs/clue-plugin-howler
cilium/docker-plugin
grafana/plugin-build-report-service
intel/device-mgmt-toolkit-rpc-go
eclipse/che-plugin-broker
starlingx/intel-fpga-plugin
rancher/harvester-lvm-csi-plugin
hashicorp/vault-plugin-secrets-gcp
rocm/device-config-manager
hashicorp/vault-plugin-secrets-kv
kong/plugin-example
hashicorp/vault-plugin-secrets-azure
rancher/hardened-sriov-network-device-plugin
starlingx/k8s-plugins-sriov-network-device
intel/gaudi-device-plugin
intel/intel-deviceplugin-operator
kasmweb/kasm-network-plugin
modular/max-nvidia-base
与「nvidia-kubevirt-gpu-device-plugin」相关的博客与命名空间
相关博客
VLLM Docker 容器化部署指南:在 NVIDIA Jetson 平台高效运行大语言模型推理服务
VLLM是一个高效的开源大语言模型(LLM)推理服务框架,通过创新的PagedAttention技术实现高吞吐量和低延迟的推理性能。本文介绍的`dustynv/vllm`镜像是针对NVIDIA Jetson平台优化的容器化版本,由[dustynv/jetson-containers](https://github.com/dustynv/jetson-containers)项目构建,专为边缘计算场景设计,支持在资源受限的嵌入式设备上部署高性能LLM推理服务。
GPUSTACK Docker 容器化部署指南(生产级规范与最佳实践)
GPUSTACK 是一款专注于 GPU 集群管理的中间件,旨在简化大语言模型(LLMs)及其他 GPU 密集型应用的部署与运行流程。通过统一的集群管理接口,GPUSTACK 能够高效调度 GPU 资源、优化任务分配,并提供监控与运维支持,适用于 AI 实验室、企业级 AI 平台等场景。
NVIDIA CUDA 镜像 Docker 容器化部署全流程
CPU 像“全能但慢的多面手”,适合处理逻辑复杂但数据量小的任务;GPU 像“成千上万的小工人”,擅长同时处理大量重复、简单的计算。CUDA 就是连接开发者与 GPU 能力的“桥梁”,让 GPU 能脱离显卡驱动,直接为科学计算、AI 训练、数据处理等任务服务。