7. Production serving
The last chapter packages your own architecture the way MAX packages its own, serves and measures it, and ends on Modular’s cloud.
| Section | You learn |
|---|---|
| 7.1 Serve your architecture | Packaging your model as a MAX architecture and serving it with max serve --custom-architectures. |
| 7.2 Benchmark a server | TTFT, TPOT, and throughput with max benchmark, for MAX’s SmolLM2 and yours. |
| 7.3 Structured output and tools | JSON that follows a schema, and function calling built on top of it. |
| 7.4 Modular’s cloud | Pointing the same client at Modular’s cloud, an inference service that runs state-of-the-art open models on MAX. |
Related handbook pages: OpenAI-compatible API, structured outputs, function calling, and performance benchmarks.