# Examples

Dedicated provisioning is used for high-throughput or latency-sensitive applications. Whether you're processing large volumes of documents or streaming real-time events, having compute resources reserved just for your workload ensures performance and control.

This section provides hands-on examples for common use cases:

#### 🔁 Batch Jobs

Run inference over thousands of documents in parallel with no rate limits. With LangChain’s `batch()` support, you can efficiently summarize, classify, or extract data at scale.

#### ⚡ Real-Time Streaming

Process live data streams (like Reddit comments or tweets) with consistent, low-latency inference. Ideal for chatbots, moderation tools, and classification pipelines.