portkey-typescript-sdk
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
ChinesePortkey TypeScript SDK
Portkey TypeScript SDK
<!--
PLACEHOLDER: Replace this content with your actual SKILL.md content.
This file should contain comprehensive instructions for AI agents on how to:
- Install and configure the Portkey TypeScript SDK
- Use core features (completions, chat, embeddings)
- Implement observability and tracing
- Set up caching, fallbacks, and load balancing
- Handle errors properly
- Work with different LLM providers through Portkey
Keep the file under 500 lines. Move detailed reference material to the
references/ directory.
-->
<!--
占位符：将此内容替换为实际的SKILL.md内容。
本文件应包含AI代理所需的全面说明，包括：
- 安装和配置Portkey TypeScript SDK
- 使用核心功能（补全、对话、嵌入）
- 实现可观测性和追踪
- 设置缓存、降级和负载均衡
- 正确处理错误
- 通过Portkey与不同的LLM提供商协作
请将文件控制在500行以内。详细参考资料请移至references/目录。
-->
When to use this skill
何时使用该技能
Use this skill when:
- The user wants to integrate Portkey into a TypeScript or JavaScript application
- The user needs LLM observability, caching, or reliability features
- The user wants to route requests across multiple LLM providers
- The user mentions "Portkey", "AI gateway", or "LLM observability" in Node.js/TS context
当以下情况时使用本技能：
- 用户希望将Portkey集成到TypeScript或JavaScript应用中
- 用户需要LLM可观测性、缓存或可靠性相关功能
- 用户希望在多个LLM提供商之间路由请求
- 用户在Node.js/TS场景中提到了“Portkey”、“AI gateway”或“LLM observability”
Installation
安装
bash
npm install portkey-aibash
npm install portkey-aior
or
yarn add portkey-ai
yarn add portkey-ai
or
or
pnpm add portkey-ai
undefinedpnpm add portkey-ai
undefinedQuick Start
快速开始
typescript
import Portkey from 'portkey-ai';
const client = new Portkey({
apiKey: 'YOUR_PORTKEY_API_KEY',
virtualKey: 'YOUR_VIRTUAL_KEY', // Optional: for provider routing
});
const response = await client.chat.completions.create({
model: 'gpt-4',
messages: [
{ role: 'user', content: 'Hello!' }
],
});
console.log(response.choices[0].message.content);typescript
import Portkey from 'portkey-ai';
const client = new Portkey({
apiKey: 'YOUR_PORTKEY_API_KEY',
virtualKey: 'YOUR_VIRTUAL_KEY', // Optional: for provider routing
});
const response = await client.chat.completions.create({
model: 'gpt-4',
messages: [
{ role: 'user', content: 'Hello!' }
],
});
console.log(response.choices[0].message.content);