70 lines
977 B
Markdown
70 lines
977 B
Markdown
---
|
|
layout: two-cols
|
|
layoutClass: gap-8
|
|
---
|
|
|
|
# Hello World con LangChain.js
|
|
|
|
Tu primera aplicación LLM
|
|
|
|
```typescript {all|1-2|4-7|9-11|13-15|all}
|
|
// index.ts
|
|
import { ChatOpenAI } from "@langchain/openai";
|
|
|
|
// 1. Crear el modelo
|
|
const model = new ChatOpenAI({
|
|
model: "gpt-4.1-mini",
|
|
});
|
|
|
|
// 2. Invocar el modelo
|
|
const response = await model.invoke(
|
|
"¿Qué es LangChain?"
|
|
);
|
|
|
|
// 3. Ver la respuesta
|
|
console.log(response.content);
|
|
```
|
|
|
|
::right::
|
|
|
|
<div class="mt-12 text-sm pl-4">
|
|
|
|
<v-click>
|
|
|
|
## 🏃 Ejecutar con tsx
|
|
|
|
```bash
|
|
# Instalar tsx
|
|
npm install -D tsx
|
|
|
|
# Ejecutar directamente
|
|
npx tsx index.ts
|
|
```
|
|
|
|
</v-click>
|
|
|
|
<v-click>
|
|
|
|
## 📤 Salida Esperada
|
|
|
|
```
|
|
LangChain es un framework para
|
|
desarrollar aplicaciones con LLMs.
|
|
Permite conectar modelos de lenguaje
|
|
con datos externos y herramientas...
|
|
```
|
|
|
|
</v-click>
|
|
|
|
<v-click>
|
|
|
|
## 🎯 Conceptos Clave
|
|
|
|
- **Model**: ChatOpenAI
|
|
- **invoke()**: Llamada síncrona
|
|
- **response.content**: El texto generado
|
|
|
|
</v-click>
|
|
|
|
</div>
|