- Server Arguments
- Session-Aware Radix Cache
- Hyperparameter Tuning
- Attention Backend
- Speculative Decoding
- Structured Outputs
- Quantization
- Expert Parallelism
- Prefill Context Parallelism
- Decode Context Parallelism
- LoRA
- PD Disaggregation
- Pipeline Parallelism
- HiCache
- SGLang Simulator
- Observability
- And more…
Advanced Features
Advanced Features
Advanced configuration, optimization, and deployment features for SGLang.
