EXAMPLES.md

🎯 Examples Hub - Quick Navigation

Bite-sized wisdom, maximum impact

Example Categories

🚀 Quick Start

• Basic Examples - Hello world to first app
• Streaming Examples - Real-time generation
• Batch Processing - Efficient bulk operations

💡 Common Use Cases

• Chat Applications - Building conversational AI
• API Servers - REST/GraphQL/WebSocket
• Document QA - RAG systems

🔧 Advanced Patterns

• Custom Routing - Smart model selection
• Caching Strategies - Performance optimization
• Model Ensemble - Multi-model inference

🏭 Production

• Docker Examples - Containerization
• Kubernetes - Orchestration at scale
• Monitoring - Observability

Quick Code Snippets

Minimal Setup (3 lines!)

import LLMRouter from 'llm-runner-router';
const router = new LLMRouter();
const response = await router.quick("Hello AI!");

Load Any Model Format

await router.load('model.gguf');     // GGUF
await router.load('model.onnx');     // ONNX
await router.load('hf:gpt2');        // HuggingFace

Stream Tokens

for await (const token of router.stream(prompt)) {
  process.stdout.write(token);
}

Custom Strategy

router.setStrategy('quality-first'); // Or: cost-optimized, speed-priority

Interactive Examples

Try these in your browser console:

// Browser-ready example
const router = new LLMRouter({ 
  preferredEngine: 'webgpu' 
});
await router.load('tinyllama.gguf');
console.log(await router.quick("Hi!"));

Learning Path

1. Start Here: Basic Examples (5 min)
2. Then: API Server (10 min)
3. Advanced: Custom Routing (15 min)
4. Production: Docker Deploy (20 min)

---

Examples are the bridges between theory and mastery 🌉