# User Guide

Use these pages as reference material while editing configs, command lines, or
deployment settings.

- [Configuration](configuration.md): engine, sampler, benchmark, and runtime
  parameter reference.
- [Models](models.md): model family, strategy, and sampling compatibility.
- [Benchmark](benchmark.md): config-first evaluation workflow.
- [Server](server.md): HTTP serving commands, request formats, and local demo
  visualization.
- [Features](features/index.md): focused pages for prefix caching, CUDA Graph,
  LoRA, parallelism, and kernels.
- [Troubleshooting](troubleshooting.md): common environment, CUDA, config, and
  serving failures.

Start with [Configuration](configuration.md) when a command line is unclear.
Use [Models](models.md) before combining a new `model_name`,
`decoding_strategy`, and `sampling_mode`.