> ## Documentation Index
> Fetch the complete documentation index at: https://code.pipellm.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Introduction

<Frame>
  <img src="https://mintcdn.com/pipellmcode/Qpc-WMmpNmVCsOeu/images/PIPELLMCover.png?fit=max&auto=format&n=Qpc-WMmpNmVCsOeu&q=85&s=38544f70244fdbe01acc997680899aa9" alt="PIPELLM Cover" width="1200" height="675" data-path="images/PIPELLMCover.png" />
</Frame>

# Introduction

In an era of diverse AI models, PIPELLM Gateway builds unified and efficient model integration infrastructure for enterprises, breaking down technical barriers and simplifying model invocation.

# Core Capabilities

* **Intelligent Routing & Distribution:** Automatically selects optimal models for efficient request handling
* **Real-time Health Monitoring:** 24/7 heartbeat detection with automatic failover
* **Enterprise-Grade Deployment:** Supports private and hybrid cloud for security compliance
* **Elastic Load Balancing:** Handles traffic spikes and ensures service stability
* **Granular Cost Control:** Real-time cost tracking to optimize AI spending
* **Unified API Standards:** One interface adapts to multiple models, reducing development complexity
* **Intelligent Caching:** Reuses similar requests to reduce costs and improve speed
* **Full-Chain Observability:** Log tracing, performance monitoring, and call analysis
* **Flexible Rate Limiting & Circuit Breaking:** Protects backend services and ensures system stability

# Core Philosophy

**"Any SDK, Any Model, No Lock-in"**

* Freely switch between any SDK and any model, never locked into a single vendor. Technology empowers business, not constrains innovation.

# Use Cases

* AI Startups: Reduce infrastructure costs
* R\&D Teams: Quickly compare and validate multiple models
* Cost-Conscious Enterprises: Intelligently route to cost-effective models
* Finance & Healthcare: Private deployment with data staying in-house

# Special Note

<Check>
  [https://pipellm.ai](https://pipellm.ai)

  Provide professional enterprise-level services, natural language models, image models, video models, model hosting, and model management.\
  Source:Google / AWS / Azure / Volcano Engine / Openrouter / Together / Fireworks / Nvidia Partner
</Check>

<Danger>
  [https://code.pipellm.ai](https://code.pipellm.ai)

  We will offer enterprise-level service surplus to users at half price for use in Code scenarios, as well as provide free support for certain public welfare projects. To ensure user stability, we will periodically close registrations to maintain stable operations.

  Source: Google / AWS **Private Price**
</Danger>