//! Expose your local LLM as an OpenAI-compatible API.
//!
//! Run with: cargo run --example serve_model --features serve
//!
//! Then test with:
//! curl http://localhost:8090/v1/models
//! curl -X POST http://localhost:8090/v1/chat/completions \
//! -H "Content-Type: application/json" \
//! -d '{"model":"llama3.2:1b","messages":[{"role":"user","content":"Hello!"}]}'
use ;
async