Back to Details

ppio-docs

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

PPIO 平台参考

PPIO Platform Reference

按以下流程处理：
  1. 识别用户所属产品方向（模型服务、Agent 沙箱、GPU 容器、集成、故障排查）。
  2. 使用 WebFetch 或等效 HTTP 读取工具，从下方映射表中获取所需的最少参考文件。
  3. 将价格、模型可用性和服务状态视为动态数据，回答前先获取最新信息。
  4. 尽量返回可直接复制粘贴的命令或代码片段。
  5. 如果本地参考文件未涵盖所需内容，提供最接近的经验证指引，明确说明缺口，并指向官方实时文档。
Follow the following process:
  1. Identify the user's product direction (Model Services, Agent Sandbox, GPU Containers, Integration, Troubleshooting).
  2. Use WebFetch or equivalent HTTP reading tools to obtain the minimum required reference files from the mapping table below.
  3. Treat pricing, model availability, and service status as dynamic data; obtain the latest information before answering.
  4. Try to return commands or code snippets that can be directly copied and pasted.
  5. If the local reference files do not cover the required content, provide the closest verified guidelines, clearly state the gaps, and point to the official real-time documentation.

LLM(大语言模型) 快速参考

LLM( Large Language Model) Quick Reference

键值
API Base URL
https://api.ppio.com/openai
鉴权方式
Authorization: Bearer <API_KEY>
获取 API Keyhttps://ppio.com/settings/key-management
模型目录https://ppio.com/ai-computing/llm-api
定价https://ppio.com/pricing
控制台https://ppio.com/console
KeyValue
API Base URL
https://api.ppio.com/openai
Authentication Method
Authorization: Bearer <API_KEY>
Obtain API Keyhttps://ppio.com/settings/key-management
Model Directoryhttps://ppio.com/ai-computing/llm-api
Pricinghttps://ppio.com/pricing
Consolehttps://ppio.com/console

文档映射表

Documentation Mapping Table

问题读取此文件实时文档
快速入门quick-start.md-
LLM API 使用llm-guide.mdAPI 参考
Agent 沙箱sandbox-guide.md文档
GPU 实例gpu-guide.md文档
工具集成integrations.md文档
常见问题faq-and-issues.mdFAQ
QuestionRead This FileReal-Time Documentation
Quick Startquick-start.md-
LLM API Usagellm-guide.mdAPI Reference
Agent Sandboxsandbox-guide.mdDocumentation
GPU Instancesgpu-guide.mdDocumentation
Tool Integrationintegrations.mdDocumentation
Frequently Asked Questionsfaq-and-issues.mdFAQ

参考文件选择规则

Reference File Selection Rules

  • 初次配置和最简示例使用 
    quick-start.md
    （见上方映射表）。
  • 聊天补全、工具调用、视觉、JSON 模式和批量任务使用 
    llm-guide.md
    （见上方映射表）。
  • SDK/CLI、生命周期、模板和文件操作使用 
    sandbox-guide.md
    （见上方映射表）。
  • 选择 GPU 实例与无服务器 GPU 以及定价权衡使用 
    gpu-guide.md
    （见上方映射表）。
  • 集成问题统一以 
    integrations.md
    （见上方映射表）为入口，再从中选择子文档。
  • 调试时优先使用 
    faq-and-issues.md
    （见上方映射表）。
  • 如果无法从 GitHub 获取原始文件，回退到本技能目录中的本地 
    references/*.md
    文件。
  • 若答案涉及配额、定价、模型数量、延迟或硬件可用性，请先通过实时文档/API 验证，并在响应中注明验证日期。
  • For initial configuration and minimal examples, use 
    quick-start.md
    (see mapping table above).
  • For chat completion, tool calling, vision, JSON mode, and batch tasks, use 
    llm-guide.md
    (see mapping table above).
  • For SDK/CLI, lifecycle, templates, and file operations, use 
    sandbox-guide.md
    (see mapping table above).
  • For selecting GPU instances vs serverless GPU and pricing trade-offs, use 
    gpu-guide.md
    (see mapping table above).
  • For integration issues, uniformly use 
    integrations.md
    (see mapping table above) as the entry point, then select sub-documents from it.
  • For debugging, prioritize using 
    faq-and-issues.md
    (see mapping table above).
  • If the raw file cannot be obtained from GitHub, fall back to the local 
    references/*.md
    files in this skill directory.
  • If the answer involves quotas, pricing, number of models, latency, or hardware availability, first verify through real-time documentation/API, and note the verification date in the response.

动态数据策略

Dynamic Data Strategy

  • 将模型目录大小、定价、限流、延迟和区域可用性视为动态数据。
  • 在最终响应中以 
    Verified on: YYYY-MM-DD
    格式标注动态数据的验证时间。
  • 如果无法实时验证，明确说明数据可能已过期，并附上官方来源链接。
  • Treat model directory size, pricing, rate limiting, latency, and regional availability as dynamic data.
  • Mark the verification time of dynamic data in the final response in the format 
    Verified on: YYYY-MM-DD
    .
  • If real-time verification is not possible, clearly state that the data may be outdated and attach the link to the official source.

动态数据（始终获取最新值）

Dynamic Data (Always Get the Latest Values)

获取最新模型列表

Get the Latest Model List

bash
curl https://api.ppio.com/openai/v1/models \
  -H "Authorization: Bearer $PPIO_API_KEY"
bash
curl https://api.ppio.com/openai/v1/models \
  -H "Authorization: Bearer $PPIO_API_KEY"

当前定价

Current Pricing

访问 https://ppio.com/pricing 查看最新定价。
Visit https://ppio.com/pricing to view the latest pricing.

更新日志

Changelog