46 lines
716 B
Markdown
46 lines
716 B
Markdown
---
|
|
layout: default
|
|
---
|
|
|
|
# Batch: Procesamiento en Lote
|
|
|
|
<div class="mt-8">
|
|
|
|
<v-click>
|
|
|
|
<div class="text-sm mb-6">
|
|
|
|
**¿Cuándo usar batch?**
|
|
|
|
- ✅ Procesar múltiples inputs a la vez
|
|
- ✅ Optimización de costos (algunos proveedores ofrecen descuentos)
|
|
- ✅ Mejor throughput que llamadas individuales
|
|
|
|
</div>
|
|
|
|
</v-click>
|
|
|
|
<v-click>
|
|
|
|
```typescript
|
|
import { ChatOpenAI } from "@langchain/openai";
|
|
|
|
const model = new ChatOpenAI({ model: "gpt-4.1-mini" });
|
|
|
|
const inputs = [
|
|
"¿Qué es un agente?",
|
|
"¿Qué es RAG?",
|
|
"¿Qué es LangSmith?"
|
|
];
|
|
|
|
const responses = await model.batch(inputs);
|
|
|
|
responses.forEach((response, i) => {
|
|
console.log(`Pregunta ${i + 1}:`, response.content);
|
|
});
|
|
```
|
|
|
|
</v-click>
|
|
|
|
</div>
|