runpod-usage
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
ChineseRunpod usage (concepts)
Runpod 使用指南(概念篇)
Background knowledge for making the right choice before you act. This skill runs
nothing — once you know what to do, execute with runpod-mcp/runpodctl
(infra), flash (your own code), or companion-clis (models/images/data).
Read the one reference file that matches the question:
| Question | Read |
|---|---|
First-run setup / auth — get + set | |
| Pods vs serverless, workers, cold starts, FlashBoot, queue vs load-balanced | |
| The development loop for ANY workload (start here) — plan → prefer prebuilt → provision → verify → teardown | |
| Stand up / iterate a workload on a pod — the pod sub-loop | |
| Deploy / iterate a serverless endpoint — Hub vs flash vs custom, invoke + verify | |
Install software on a pod — package hygiene, | |
Build a Docker image Runpod can run (handler contract, Dockerfile, | |
| How to build an image well — base image, layering, bake-in vs volume, pod vs serverless (queue/LB) contract | |
| Where data lives — container disk vs network volume, model caching, S3 access | |
| Which GPU / how much VRAM / cost & availability / data centers | |
| Reaching a pod or endpoint over HTTP (proxy URLs, exposed ports) | |
| Common mistakes and how to avoid them | |
这是你采取行动前做出正确选择所需的背景知识。本技能不执行任何操作——当你明确操作步骤后,可通过runpod-mcp/runpodctl(基础设施)、flash(自定义代码)或companion-clis(模型/镜像/数据)执行操作。
请阅读与问题匹配的参考文档:
| 问题 | 参考文档 |
|---|---|
首次运行设置/认证——获取并配置 | |
| Pods与Serverless对比、工作节点、冷启动、FlashBoot、队列 vs 负载均衡 | |
| 任意工作负载的开发循环(从这里开始)——规划→优先使用预构建镜像→资源配置→验证→销毁 | |
| 在Pod上部署/迭代工作负载——Pod子循环 | |
| 部署/迭代Serverless端点——Hub vs flash vs 自定义、调用+验证 | |
在Pod上安装软件——包管理规范、 | |
构建Runpod可运行的Docker镜像(处理器合约、Dockerfile、 | |
| 如何构建优质镜像——基础镜像、分层、内置 vs 卷、Pod与Serverless(队列/负载均衡)合约 | |
| 数据存储位置——容器磁盘 vs 网络卷、模型缓存、S3访问 | |
| 选择哪款GPU / 需要多少VRAM / 成本与可用性 / 数据中心 | |
| 通过HTTP访问Pod或端点(代理URL、暴露端口) | |
| 常见错误及规避方法 | |