# SGLang Documentation

## Docs

- [Welcome to SGLang](https://lmsysorg-dsv4-1.mintlify.site/index.md): High-performance serving framework for large language and multimodal models.
- [Installation](https://lmsysorg-dsv4-1.mintlify.site/docs/get-started/install.md): Install SGLang with pip/uv, source, Docker, Kubernetes, and cloud deployment options.
- [Quickstart](https://lmsysorg-dsv4-1.mintlify.site/docs/get-started/quickstart.md): Get up and running with SGLang in minutes: install, launch a server, and send your first request.
- [Tutorial: Sending a request](https://lmsysorg-dsv4-1.mintlify.site/docs/basic_usage/send_request.md)
- [Basic Usage](https://lmsysorg-dsv4-1.mintlify.site/docs/basic_usage/overview.md): Core APIs and common usage patterns for SGLang.
- [OpenAI-Compatible APIs](https://lmsysorg-dsv4-1.mintlify.site/docs/basic_usage/openai_api.md): Documentation for OpenAI-Compatible APIs
- [OpenAI APIs - Completions](https://lmsysorg-dsv4-1.mintlify.site/docs/basic_usage/openai_api_completions.md)
- [OpenAI APIs - Vision](https://lmsysorg-dsv4-1.mintlify.site/docs/basic_usage/openai_api_vision.md)
- [OpenAI APIs - Embedding](https://lmsysorg-dsv4-1.mintlify.site/docs/basic_usage/openai_api_embeddings.md)
- [Anthropic-Compatible API](https://lmsysorg-dsv4-1.mintlify.site/docs/basic_usage/anthropic_api.md): Use the Anthropic Messages API (/v1/messages) with SGLang, including Claude Code integration and prefix-cache tuning.
- [Ollama-Compatible API](https://lmsysorg-dsv4-1.mintlify.site/docs/basic_usage/ollama_api.md)
- [Offline Engine API](https://lmsysorg-dsv4-1.mintlify.site/docs/basic_usage/offline_engine_api.md)
- [SGLang Native APIs](https://lmsysorg-dsv4-1.mintlify.site/docs/basic_usage/native_api.md)
- [Sampling Parameters](https://lmsysorg-dsv4-1.mintlify.site/docs/basic_usage/sampling_params.md)
- [Advanced Features](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/overview.md): Advanced configuration, optimization, and deployment features for SGLang.
- [Server Arguments](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/server_arguments.md)
- [Session-Aware Radix Cache](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/session_radix_cache.md)
- [Radix Cache Eviction Policies](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/radix_eviction_policy.md)
- [Hyperparameter Tuning](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/hyperparameter_tuning.md)
- [Attention Backend](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/attention_backend.md)
- [HiSparse: Hierarchical Sparse Attention](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/hisparse_guide.md)
- [Speculative Decoding](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/speculative_decoding.md)
- [Adaptive Speculative Decoding](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/adaptive_speculative_decoding.md)
- [Structured Outputs](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/structured_outputs.md)
- [Structured Outputs For Reasoning Models](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/structured_outputs_for_reasoning_models.md)
- [Tool Parser](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/tool_parser.md)
- [Reasoning Parser](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/separate_reasoning.md)
- [Quantization](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/quantization.md)
- [Quantized KV Cache](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/quantized_kv_cache.md)
- [Reasoning-Aware Compression](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/reasoning_aware_compression.md)
- [DP, DPA and SGLang DP Router](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/dp_dpa_smg_guide.md)
- [Decode Context Parallelism](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/dcp.md)
- [Expert Parallelism](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/expert_parallelism.md)
- [LoRA Serving](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/lora.md)
- [PD Disaggregation](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/pd_disaggregation.md)
- [EPD Disaggregation](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/epd_disaggregation.md)
- [Pipeline Parallelism for Long Context](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/pipeline_parallelism.md)
- [Hierarchical KV Caching (HiCache)](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/hicache.md)
- [SGLang HiCache Best Practices](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/hicache_best_practices.md)
- [HiCache System Design and Optimization](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/hicache_design.md)
- [Runtime Attach/Detach HiCache Storage Backend (No Restart)](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/hicache_storage_runtime_attach_detach.md)
- [SGLang Simulator](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/sglang_simulator.md)
- [Query VLM with Offline Engine](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/vlm_query.md)
- [DP for Multi-Modal Encoder in SGLang](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/dp_for_multi_modal_encoder.md)
- [Cuda Graph for Multi-Modal Encoder in SGLang](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/cuda_graph_for_multi_modal_encoder.md)
- [Breakable CUDA Graph](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/breakable_cuda_graph.md)
- [Piecewise CUDA Graph](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/piecewise_cuda_graph.md)
- [SGLang Model Gateway](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/sgl_model_gateway.md)
- [llm-d](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/llm-d.md)
- [Deterministic Inference](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/deterministic_inference.md)
- [Observability](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/observability.md)
- [Model Loading](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/model_loading.md): Control how SGLang loads model weights: load formats, model loader extra config, multithreaded loading, prefetching, and remote/streaming loaders.
- [Loading Models from Object Storage](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/object_storage.md)
- [Checkpoint Engine Integration](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/checkpoint_engine.md)
- [SGLang for RL Systems](https://lmsysorg-dsv4-1.mintlify.site/docs/advanced_features/sglang_for_rl.md)
- [Supported models](https://lmsysorg-dsv4-1.mintlify.site/docs/supported-models.md): See which families of SGLang-compatible models are actively maintained.
- [Large Language Models](https://lmsysorg-dsv4-1.mintlify.site/docs/supported-models/generative_models.md)
- [Multimodal Language Models](https://lmsysorg-dsv4-1.mintlify.site/docs/supported-models/multimodal_language_models.md)
- [Diffusion language models](https://lmsysorg-dsv4-1.mintlify.site/docs/supported-models/diffusion_language_models.md)
- [Embedding models](https://lmsysorg-dsv4-1.mintlify.site/docs/supported-models/embedding_models.md): Dense and sparse embedding models with FlashInfer acceleration and SGLang's batching infrastructure.
- [Rerank models](https://lmsysorg-dsv4-1.mintlify.site/docs/supported-models/rerank_models.md)
- [Classification Models](https://lmsysorg-dsv4-1.mintlify.site/docs/supported-models/classify_models.md)
- [Reward models](https://lmsysorg-dsv4-1.mintlify.site/docs/supported-models/reward_models.md)
- [How to Support New Models](https://lmsysorg-dsv4-1.mintlify.site/docs/supported-models/support_new_models.md): This document explains how to add support for new language models and multimodal large language models (MLLMs) in SGLang. It also covers how to test new models and register external implementations.
- [Transformers Fallback in SGLang](https://lmsysorg-dsv4-1.mintlify.site/docs/supported-models/transformers_fallback.md)
- [Use Models From ModelScope](https://lmsysorg-dsv4-1.mintlify.site/docs/supported-models/modelscope.md)
- [MindSpore Models](https://lmsysorg-dsv4-1.mintlify.site/docs/supported-models/mindspore_models.md)
- [Developer Guide](https://lmsysorg-dsv4-1.mintlify.site/docs/developer_guide/overview.md): Contributing to SGLang — development setup, benchmarking, and evaluation.
- [Contribution Guide](https://lmsysorg-dsv4-1.mintlify.site/docs/developer_guide/contribution_guide.md)
- [Add an out-of-tree serve backend](https://lmsysorg-dsv4-1.mintlify.site/docs/developer_guide/serve_backend_plugins.md): Connect an ecosystem runtime to sglang serve through the versioned serve backend plugin API.
- [Development Guide Using Docker](https://lmsysorg-dsv4-1.mintlify.site/docs/developer_guide/development_guide_using_docker.md)
- [Development Guide for JIT Kernels](https://lmsysorg-dsv4-1.mintlify.site/docs/developer_guide/development_jit_kernel_guide.md)
- [Quantization Contribution Guide](https://lmsysorg-dsv4-1.mintlify.site/docs/developer_guide/quantization_contribution_guide.md)
- [Benchmark and Profiling](https://lmsysorg-dsv4-1.mintlify.site/docs/developer_guide/benchmark_and_profiling.md)
- [Bench Serving Guide](https://lmsysorg-dsv4-1.mintlify.site/docs/developer_guide/bench_serving.md)
- [Evaluating New Models with SGLang](https://lmsysorg-dsv4-1.mintlify.site/docs/developer_guide/evaluating_new_models.md)
- [MSProbe Debugging Guide](https://lmsysorg-dsv4-1.mintlify.site/docs/developer_guide/msprobe_debugging_guide.md)
- [References](https://lmsysorg-dsv4-1.mintlify.site/docs/references/overview.md): FAQ, environment variables, production metrics, deployment guides, and more.
- [Troubleshooting and Frequently Asked Questions](https://lmsysorg-dsv4-1.mintlify.site/docs/references/faq.md)
- [Environment Variables](https://lmsysorg-dsv4-1.mintlify.site/docs/references/environment_variables.md)
- [Production Metrics](https://lmsysorg-dsv4-1.mintlify.site/docs/references/production_metrics.md)
- [Production Request Tracing](https://lmsysorg-dsv4-1.mintlify.site/docs/references/production_request_trace.md)
- [Multi-Node Deployment](https://lmsysorg-dsv4-1.mintlify.site/docs/references/multi_node_deployment/multi_node_index.md)
- [Multi-Node Deployment](https://lmsysorg-dsv4-1.mintlify.site/docs/references/multi_node_deployment/multi_node.md)
- [Deploy On Kubernetes](https://lmsysorg-dsv4-1.mintlify.site/docs/references/multi_node_deployment/deploy_on_k8s.md)
- [LWS Based PD Deploy](https://lmsysorg-dsv4-1.mintlify.site/docs/references/multi_node_deployment/lws_pd/lws_pd_deploy.md)
- [DeepSeekV32-Exp RBG Based PD Deploy](https://lmsysorg-dsv4-1.mintlify.site/docs/references/multi_node_deployment/rbg_pd/deepseekv32_pd.md)
- [Amazon SageMaker AI](https://lmsysorg-dsv4-1.mintlify.site/docs/basic_usage/aws_sagemaker.md): Deploy SGLang on Amazon SageMaker AI endpoints using the AWS Deep Learning Container.
- [Custom Chat Template](https://lmsysorg-dsv4-1.mintlify.site/docs/references/custom_chat_template.md)
- [Frontend Language](https://lmsysorg-dsv4-1.mintlify.site/docs/references/frontend/frontend_index.md)
- [SGLang Frontend Language](https://lmsysorg-dsv4-1.mintlify.site/docs/references/frontend/frontend_tutorial.md)
- [Choices Methods in SGLang](https://lmsysorg-dsv4-1.mintlify.site/docs/references/frontend/choices_methods.md)
- [Server Arguments](https://lmsysorg-dsv4-1.mintlify.site/cookbook/base/reference/server_arguments.md)
- [Post-Training Integration](https://lmsysorg-dsv4-1.mintlify.site/docs/references/post_training_integration.md)
- [Nightly precision regression](https://lmsysorg-dsv4-1.mintlify.site/docs/references/nightly_precision_regression.md)
- [Hardware Platforms](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/overview.md): Platform-specific guides for running SGLang on GPUs, TPUs, NPUs, CPUs, and more.
- [NVIDIA GPUs](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/nvidia-gpus.md)
- [AMD GPUs](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/amd_gpu.md)
- [Apple Silicon with Metal](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/apple_metal.md)
- [Quickstart](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/getting-started/quick_start.md)
- [Installation](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/getting-started/installation.md): Complete installation guide for SGLang on Ascend NPUs, including component version mapping, environment setup, and launching inference services.
- [Supported Models](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/reference/support_models.md)
- [Supported Features](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/reference/support_features.md)
- [Environment Variables](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/reference/environment_variables.md)
- [Glossary](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/reference/glossary.md)
- [DeepSeek-R1](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/tutorials/deepseek_r1.md)
- [DeepSeek-V3.2](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/tutorials/deepseek_v3_2.md)
- [DeepSeek-V4-Flash](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/tutorials/deepseek_v4_flash.md)
- [GLM-5.1](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/tutorials/glm_5_1.md)
- [GLM-5.2](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/tutorials/glm_5_2.md)
- [Hy3](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/tutorials/hy3.md)
- [Kimi-K2.6](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/tutorials/kimi_k2_6.md)
- [MiniMax-M2.5](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/tutorials/minimax_m2_5.md)
- [MiMo-V2-Flash](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/tutorials/mimo_v2_flash.md)
- [Qwen3-8B](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/tutorials/qwen3_8b.md)
- [Qwen3-32B](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/tutorials/qwen3_32b.md)
- [Qwen3-30B-A3B](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/tutorials/qwen3_30b_a3b.md)
- [Qwen3-235B-A22B](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/tutorials/qwen3_235b_a22b.md)
- [Qwen3.5-397B-A17B](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/tutorials/qwen3_5_397b.md)
- [Qwen3.6-27B](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/tutorials/qwen3_6_27b.md)
- [Qwen3.6-35B-A3B](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/tutorials/qwen3_6_35b_a3b.md)
- [Qwen3-Next-80B-A3B-Instruct](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/tutorials/qwen3_next_80b_a3b_instruct.md)
- [Qwen3.8-Max](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/tutorials/qwen3_8_max.md)
- [DeepSeek-R1](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/best-practices/deepseek_r1.md)
- [DeepSeek-V3.2](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/best-practices/deepseek_v3_2.md)
- [DeepSeek-V4-Flash](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/best-practices/deepseek_v4_flash.md)
- [GLM-5.1](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/best-practices/glm_5_1.md)
- [GLM-5.2](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/best-practices/glm_5_2.md)
- [Kimi-K2.6](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/best-practices/kimi_k2_6.md)
- [MiniMax-M2.5](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/best-practices/minimax_m2_5.md)
- [MiMo-V2-Flash](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/best-practices/mimo_v2_flash.md)
- [Qwen3-8B](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/best-practices/qwen3_8b.md)
- [Qwen3-32B](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/best-practices/qwen3_32b.md)
- [Qwen3-30B-A3B](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/best-practices/qwen3_30b_a3b.md)
- [Qwen3-235B-A22B](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/best-practices/qwen3_235b_a22b.md)
- [Qwen3.5-397B-A17B](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/best-practices/qwen3_5_397b.md)
- [Qwen3.6-27B](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/best-practices/qwen3_6_27b.md)
- [Qwen3.6-35B-A3B](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/best-practices/qwen3_6_35b_a3b.md)
- [Qwen3-Next-80B-A3B-Instruct](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/model-deployment/best-practices/qwen3_next_80b_a3b_instruct.md)
- [Accuracy Evaluation](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/evaluation/accuracy_evaluation.md)
- [Performance Testing](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/evaluation/performance_testing.md)
- [Parameter Tuning](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/optimization/parameter_tuning.md)
- [Performance Profiling](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/optimization/profiling.md)
- [Quantization on Ascend](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/optimization/quantization.md)
- [Ring-SP Performance (Wan2.1-T2V-1.3B)](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/optimization/ring_sp_performance.md)
- [Contribution Guide](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/development/contribution_guide.md)
- [How to Support New Models](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/development/support_new_models.md)
- [Operator Development](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/development/operator_development.md): How to develop custom operators (Ascend C / Triton) for Ascend NPU and integrate them into the SGLang inference engine.
- [Operator Performance Optimization](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/development/operator_performance_optimizing.md)
- [Mindspore backend](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/mindspore_backend.md)
- [Troubleshooting and FAQ](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/ascend-npus/faq.md)
- [CPU Servers](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/cpu_server.md)
- [NVIDIA Jetson Orin](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/nvidia_jetson.md): Guide for installing and running SGLang on NVIDIA Jetson Orin devices.
- [Moore Threads GPUs](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/mthreads_gpu.md)
- [TPU](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/tpu.md): SGLang supports high-performance TPU inference through the SGLang-JAX backend, which is specifically optimized for Google Cloud TPUs. The JAX-based implementation delivers exceptional throughput and low latency for Large Language Model (LLM) serving workloads on TPU hardware.
- [XPU](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/xpu.md)
- [SGLang Plugin System](https://lmsysorg-dsv4-1.mintlify.site/docs/hardware-platforms/plugin.md)
- [SGLang Cookbook](https://lmsysorg-dsv4-1.mintlify.site/cookbook/intro.md)
- [Overview](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/intro.md): Practical guides for deploying and using large language models and vision language models with SGLang.
- [K2 Horizon](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/IFM/K2-Horizon.md): Deploy the IFM K2 Horizon family with SGLang — BF16 recipes for six dense and Mixture-of-Experts checkpoints on NVIDIA H200 GPUs.
- [Kimi-K3](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Moonshotai/Kimi-K3.md): Deploy Moonshot AI's Kimi-K3 with SGLang — a 2.8T-parameter hybrid Mixture-of-Experts vision-language model (Kimi Delta Attention + MLA, 16/896 active experts) with NVIDIA, AMD, and NPU recipes.
- [Kimi-K2.7-Code](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Moonshotai/Kimi-K2.7-Code.md): Deploy Kimi-K2.7-Code with SGLang for coding-focused agentic workflows, thinking output, tool calling, and multimodal input.
- [Kimi-K2.6](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Moonshotai/Kimi-K2.6.md)
- [Kimi-K2.5](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Moonshotai/Kimi-K2.5.md)
- [Kimi-K2](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Moonshotai/Kimi-K2.md)
- [Kimi-Linear](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Moonshotai/Kimi-Linear.md)
- [Inkling](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/ThinkingMachines/Inkling.md): Deploy Inkling with SGLang — verified launch commands, tuning, and multimodal / reasoning / tool-calling usage for Thinking Machines' 975B Mixture-of-Experts model with 1M-token context.
- [Inkling-Small](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/ThinkingMachines/Inkling-Small.md): Deploy Inkling-Small with SGLang — launch commands, tuning, and multimodal / reasoning / tool-calling usage for Thinking Machines' Inkling-Small Mixture-of-Experts model.
- [GLM-5.3](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/GLM/GLM-5.3.md): Deploy GLM-5.3 with SGLang — Z.ai's DeepSeek-Sparse-Attention (DSA) Mixture-of-Experts model with MTP speculative decoding and 1M context, on H200, B200, B300, GB300, and AMD MI300X/MI325X/MI355X.
- [GLM-5.3-Flash](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/GLM/GLM-5.3-Flash.md): Deploy GLM-5.3-Flash with SGLang using recipes for H100, H200, B200, B300, GB200, and GB300, with MTP and multimodal serving.
- [GLM-5.2](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/GLM/GLM-5.2.md): Deploy GLM-5.2 with SGLang — Z.ai's DeepSeek-Sparse-Attention (DSA) Mixture-of-Experts model with MTP speculative decoding and 1M context, on H200, B200, B300, GB300, and AMD MI300X/MI325X/MI355X.
- [GLM-5.1](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/GLM/GLM-5.1.md)
- [GLM-5](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/GLM/GLM-5.md)
- [GLM-OCR](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/GLM/GLM-OCR.md)
- [GLM Glyph](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/GLM/GLM-Glyph.md)
- [GLM-4.7](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/GLM/GLM-4.7.md)
- [GLM-4.7-Flash](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/GLM/GLM-4.7-Flash.md)
- [GLM-4.6](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/GLM/GLM-4.6.md)
- [GLM-4.6V](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/GLM/GLM-4.6V.md)
- [GLM-4.5](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/GLM/GLM-4.5.md)
- [GLM-4.5V](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/GLM/GLM-4.5V.md)
- [Qwen3.8-Flash-Next](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Qwen/Qwen3.8-Flash-Next.md): Deploy Qwen3.8-Flash-Next with SGLang — day-0 recipes for Qwen's 176B-parameter (6B active) GDN + QSA hybrid Mixture-of-Experts preview of the Qwen4 architecture, on NVIDIA and AMD.
- [Qwen3.8](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Qwen/Qwen3.8.md): Deploy Qwen3.8 with SGLang — day-0 recipes for Qwen's 2.4T-parameter (95B active) hybrid GDN/GQA Mixture-of-Experts model on NVIDIA and AMD.
- [Qwen3.8-27B](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Qwen/Qwen3.8-27B.md): Deploy Qwen3.8-27B with SGLang — dense hybrid GDN vision-language model with BF16/FP8/NVFP4 W4A4 checkpoints and in-checkpoint MTP, single-GPU on H200, RTX PRO 6000, RTX 5090 and DGX Spark.
- [Qwen3.6](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Qwen/Qwen3.6.md)
- [Qwen3.5](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Qwen/Qwen3.5.md)
- [Qwen3](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Qwen/Qwen3.md)
- [Qwen3-Next](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Qwen/Qwen3-Next.md)
- [Qwen3-Coder](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Qwen/Qwen3-Coder.md)
- [Qwen3-Coder-Next](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Qwen/Qwen3-Coder-Next.md)
- [Qwen3-VL](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Qwen/Qwen3-VL.md)
- [Qwen2.5-VL](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Qwen/Qwen2.5-VL.md)
- [DeepSeek-V4.1](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/DeepSeek/DeepSeek-V4_1.md): Deploy DeepSeek-V4.1 Flash with SGLang — launch recipes, feature compatibility, and tuning notes for GB300, H200, B200, B300 and MI350X.
- [DeepSeek-V4](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/DeepSeek/DeepSeek-V4.md): Deploy DeepSeek-V4 with SGLang — verified launch commands, benchmarks, and tuning for Flash Official (0731), Flash, Flash Vision (Exp), Pro, and Pro Official (0813).
- [DeepSeek-V3.2](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/DeepSeek/DeepSeek-V3_2.md)
- [DeepSeek-V3.1](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/DeepSeek/DeepSeek-V3_1.md)
- [DeepSeek-V3](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/DeepSeek/DeepSeek-V3.md)
- [DeepSeek-R1](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/DeepSeek/DeepSeek-R1.md)
- [DeepSeek-Math-V2](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/DeepSeek/DeepSeek-Math-V2.md)
- [DeepSeek-OCR](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/DeepSeek/DeepSeek-OCR.md)
- [DeepSeek-OCR-2](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/DeepSeek/DeepSeek-OCR-2.md)
- [Ornith-1.0](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/DeepReinforce/Ornith-1.0.md): Deploy DeepReinforce Ornith-1.0 with SGLang - a self-improving agentic-coding model family with 397B, 35B, and 9B checkpoints plus FP8 and GGUF variants.
- [Muse Glimmer](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Meta/MuseGlimmer.md): A multimodal reasoning model served from a BF16, NVFP4 + MXFP8, vendor GGUF, or MLX checkpoint.
- [Llama 4](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Meta/Llama4.md)
- [Llama-3.3-70B](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Meta/Llama3.3-70B.md)
- [Llama-3.1](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Meta/Llama3.1.md)
- [LongCat-2.0](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Meituan/LongCat-2.0.md): Deploy LongCat-2.0-FP8 with SGLang - config-driven recipes for Meituan's 1.6T sparse MoE model on B300, B200, H200, and H20 GPUs.
- [Dots3-Note](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/RedNote/Dots3-Note.md): Deploy RedNote dots3.note with SGLang — a native multimodal omni model (MoE ViT + Whisper-derived audio encoder + native video flattening) on the dots3 hybrid MLA/SWA language model, with DSA and full-sharing MTP speculative decoding.
- [EmbeddingGemma](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Google/EmbeddingGemma.md): Serve Google's EmbeddingGemma text embedding model with SGLang.
- [Gemma 4](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Google/Gemma4.md)
- [DiffusionGemma](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Google/DiffusionGemma.md)
- [Granite 4.2](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/IBM/Granite-4.2.md): Deploy Granite 4.2 3B, 8B, and 30B dense models with SGLang on NVIDIA H200 and B200, including thinking modes and structured tool calling.
- [LFM2.5](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/LiquidAI/LFM2.5.md): Deploy Liquid AI's LFM2.5 with SGLang — hybrid gated short conv + GQA models from 350M to the 8B-A1B MoE, plus LFM2.5-VL vision, with reasoning and Pythonic tool calling.
- [GPT-OSS](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/OpenAI/GPT-OSS.md)
- [MiniMax-M3](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/MiniMax/MiniMax-M3.md): Deploy MiniMax-M3 with SGLang — a ~428B-param (23B activated) multimodal Mixture-of-Experts reasoning model with MiniMax Sparse Attention and 1M context, MXFP8 on NVIDIA Blackwell & AMD Instinct, bf16 on Hopper.
- [MiniMax-M2.7](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/MiniMax/MiniMax-M2.7.md)
- [MiniMax-M2](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/MiniMax/MiniMax-M2.md)
- [MiniMax-M2.5](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/MiniMax/MiniMax-M2.5.md)
- [Nemotron3.5-Lightning](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/NVIDIA/Nemotron3.5-Lightning.md): Deploy NVIDIA Nemotron 3.5 Lightning with SGLang — NVFP4 serving with MTP, DFlash, and DSpark speculative decoding, reasoning, and tool calling.
- [NVIDIA Nemotron3-Ultra](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/NVIDIA/Nemotron3-Ultra.md): Deploy NVIDIA Nemotron3-Ultra with SGLang - 550B hybrid MoE model (55B active) with 1M context window, BF16/NVFP4 support, built for long-running autonomous agents.
- [Nemotron 3 Nano Omni](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/NVIDIA/Nemotron3-Nano-Omni.md)
- [Nemotron3-Nano](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/NVIDIA/Nemotron3-Nano.md)
- [NVIDIA Nemotron3-Super](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/NVIDIA/Nemotron3-Super.md)
- [PaddleOCR-VL](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Baidu/PaddleOCR-VL.md): Deploy PaddleOCR-VL 1.6 / 1.5 / 0.9B with SGLang — Baidu's 0.9B NaViT + ERNIE-4.5 document-parsing VLM for OCR, tables, formulas and charts in 109 languages, on a single H100, H200 or B200.
- [Unlimited-OCR](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Baidu/Unlimited-OCR.md): Deploy Baidu Unlimited-OCR with SGLang for long document OCR using prefill-aware sliding-window attention.
- [Ernie4.5](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Ernie/Ernie4.5.md)
- [Ernie4.5-VL](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Ernie/Ernie4.5-VL.md)
- [Step-3.7-Flash (new)](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/StepFun/Step-3.7-Flash.md)
- [Step-3.5-Flash](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/StepFun/Step3.5.md)
- [Step3-VL-10B](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/StepFun/Step3-VL-10B.md)
- [Ling-3.0-flash-VL](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/InclusionAI/Ling-3.0-flash-VL.md): Deploy Ling-3.0-flash-VL with SGLang — a ~125B hybrid KDA + MLA MoE vision-language model (5.1B active) with image and video input, thinking mode, and 128K context on Blackwell and Hopper GPUs.
- [Ling-3.0-flash](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/InclusionAI/Ling-3.0-flash.md): Deploy Ling-3.0-flash with SGLang — a 124B hybrid KDA + MLA MoE in BF16, FP8, INT4, or MXFP4 on Hopper and Blackwell GPUs.
- [Ling-3.0-tiny](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/InclusionAI/Ling-3.0-tiny.md): Deploy Ling-3.0-tiny with SGLang — a compact ~7.9B total / ~1.2B active hybrid KDA + MLA MoE in BF16, FP8, or INT4, with thinking mode and tool calling.
- [Ring-2.6-1T](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/InclusionAI/Ring-2.6-1T.md)
- [Ling-2.6](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/InclusionAI/Ling-2.6.md)
- [Ling-2.5-1T](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/InclusionAI/Ling-2.5-1T.md)
- [Ring-2.5-1T](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/InclusionAI/Ring-2.5-1T.md)
- [LLaDA 2.1](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/InclusionAI/LLaDA-2.1.md)
- [Intern-S2-Mobius](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/InternLM/Intern-S2-Mobius.md): Deploy the BF16 and FP8 Intern-S2-Mobius checkpoints with SGLang on NVIDIA H200 and B200.
- [Intern-S2-Preview](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/InternLM/Intern-S2-Preview.md)
- [Intern-S1](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/InternLM/Intern-S1.md)
- [InternVL3.5](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/InternVL/InternVL3.5.md)
- [MiniCPM5-2B](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/OpenBMB/MiniCPM5-2B.md): Deploy OpenBMB MiniCPM5-2B with SGLang — a 2.5B dense Llama-architecture on-device model with 131K context, thinking mode, XML tool calling and DSpark speculative decoding on H200, RTX PRO 6000, RTX 5090 and DGX Spark.
- [MiniCPM-V 4.6](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/OpenBMB/MiniCPM-V-4_6.md)
- [Jina-reranker-m0](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Jina/Jina-reranker-m0.md)
- [Ministral-3](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Mistral/Ministral-3.md)
- [Mistral Small 4](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Mistral/Mistral-Small-4.md)
- [Mistral Medium 3.5](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Mistral/Mistral-Medium-3.5.md)
- [Devstral 2 (Mistral)](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Mistral/Devstral-2.md)
- [MiMo-V2.5](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Xiaomi/MiMo-V2.5.md)
- [MiMo-V2-Flash](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Xiaomi/MiMo-V2-Flash.md)
- [Chroma-1.0](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/FlashLabs/Chroma1.0.md)
- [Hy4 preview](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Tencent/Hy4-Preview.md): Deploy Tencent Hy4 preview with SGLang — launch recipes for the 770B-parameter Mixture-of-Experts model with MLA, DeepSeek Sparse Attention (DSA), and MTP speculative decoding, in BF16 on H200/B200/B300/GB300 and MXFP8 on Blackwell GPUs.
- [Hy3](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Tencent/Hy3.md): Deploy Tencent Hy3 with SGLang — verified launch commands and tuning for the BF16 Mixture-of-Experts model with hybrid thinking, native tool calling, 256K context, and MTP speculative decoding.
- [Hy3 preview](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Tencent/Hunyuan3-Preview.md)
- [Laguna-S-2.1](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Poolside/Laguna-S-2.1.md): Deploy poolside's Laguna-S-2.1 — a 118B hybrid-SWA Mixture-of-Experts model (8B active) for agentic coding — with SGLang on NVIDIA H200, B300, and GB300 in BF16, FP8, NVFP4, and INT4.
- [Laguna-XS-2.1](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Poolside/Laguna-XS-2.1.md): Deploy poolside's Laguna-XS-2.1 — a 33B hybrid-SWA Mixture-of-Experts model (3B active) for agentic coding — with SGLang on NVIDIA H200, B300, and GB300 in BF16, FP8, NVFP4, and INT4.
- [Laguna-M.1](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Poolside/Laguna-M.1.md): Deploy poolside's Laguna-M.1 — a 225B-parameter Mixture-of-Experts model (23B active) for agentic coding — with SGLang on NVIDIA H200, B200, B300, GB200, and GB300, across BF16, FP8, and NVFP4.
- [Laguna-XS.2](https://lmsysorg-dsv4-1.mintlify.site/cookbook/autoregressive/Poolside/Laguna-XS.2.md)
- [Overview](https://lmsysorg-dsv4-1.mintlify.site/cookbook/diffusion/intro.md): Practical guides for deploying and using diffusion models with SGLang.
- [Cosmos3](https://lmsysorg-dsv4-1.mintlify.site/cookbook/diffusion/Cosmos/Cosmos3.md)
- [FLUX](https://lmsysorg-dsv4-1.mintlify.site/cookbook/diffusion/FLUX/FLUX.md)
- [Ideogram 4](https://lmsysorg-dsv4-1.mintlify.site/cookbook/diffusion/Ideogram/Ideogram4.md)
- [Wan2.1](https://lmsysorg-dsv4-1.mintlify.site/cookbook/diffusion/Wan/Wan2.1.md)
- [Wan2.2](https://lmsysorg-dsv4-1.mintlify.site/cookbook/diffusion/Wan/Wan2.2.md)
- [LongLive 2.0](https://lmsysorg-dsv4-1.mintlify.site/cookbook/diffusion/LongLive/LongLive-2.0.md): Serve LongLive 2.0 distilled text-to-video and image-to-video models with SGLang-diffusion.
- [LTX2 & LTX2.3](https://lmsysorg-dsv4-1.mintlify.site/cookbook/diffusion/LTX/LTX2 & LTX2.3.md): Run LTX-2 and LTX-2.3 video generation pipelines with SGLang Diffusion.
- [LTX2.5](https://lmsysorg-dsv4-1.mintlify.site/cookbook/diffusion/LTX/LTX2.5.md): Run LTX-2.5 video + audio generation with SGLang Diffusion.
- [JoyAI-Echo](https://lmsysorg-dsv4-1.mintlify.site/cookbook/diffusion/JoyEcho/JoyEcho.md): Run JoyAI-Echo multi-shot audio–video generation with SGLang Diffusion.
- [Qwen-Image](https://lmsysorg-dsv4-1.mintlify.site/cookbook/diffusion/Qwen-Image/Qwen-Image.md)
- [Qwen-Image-Edit-2511](https://lmsysorg-dsv4-1.mintlify.site/cookbook/diffusion/Qwen-Image/Qwen-Image-Edit.md)
- [SenseNova-U1.5-8B-MoT](https://lmsysorg-dsv4-1.mintlify.site/cookbook/diffusion/SenseNova/SenseNova-U1.5-8B-MoT.md)
- [LongCat-Image](https://lmsysorg-dsv4-1.mintlify.site/cookbook/diffusion/LongCat/LongCat-Image.md)
- [Z-Image-Turbo](https://lmsysorg-dsv4-1.mintlify.site/cookbook/diffusion/Z-Image/Z-Image-Turbo.md)
- [Krea-2](https://lmsysorg-dsv4-1.mintlify.site/cookbook/diffusion/Krea/Krea-2.md)
- [ERNIE-Image](https://lmsysorg-dsv4-1.mintlify.site/cookbook/diffusion/Ernie-Image/Ernie-Image.md)
- [MOVA](https://lmsysorg-dsv4-1.mintlify.site/cookbook/diffusion/MOVA/MOVA.md)
- [SANA-Video](https://lmsysorg-dsv4-1.mintlify.site/cookbook/diffusion/SANA-Video/SANA-Video.md): Serve the native SANA-Video 2B 480p text-to-video model with SGLang Diffusion.
- [MiniMax-H3](https://lmsysorg-dsv4-1.mintlify.site/cookbook/diffusion/MiniMax/MiniMax-H3.md): Run native MiniMax-H3 video-and-audio generation with SGLang Diffusion.
- [LingBot Video MoE](https://lmsysorg-dsv4-1.mintlify.site/cookbook/diffusion/LingBot-Video/LingBot-Video-MoE.md): Serve the native LingBot Video MoE 30B-A3B text-to-video model with SGLang Diffusion.
- [LingBot World](https://lmsysorg-dsv4-1.mintlify.site/cookbook/diffusion/LingBot-World/LingBot-World.md)
- [LingBot World 2.0](https://lmsysorg-dsv4-1.mintlify.site/cookbook/diffusion/LingBot-World/LingBot-World-2.0.md)
- [SANA-WM](https://lmsysorg-dsv4-1.mintlify.site/cookbook/diffusion/SANA-WM/SANA-WM.md)
- [Overview](https://lmsysorg-dsv4-1.mintlify.site/cookbook/vla/intro.md): Practical guides for deploying and using Vision-Language-Action policies with SGLang.
- [Pi0.5](https://lmsysorg-dsv4-1.mintlify.site/cookbook/vla/OpenPI/Pi0.5.md)
- [Supported Models](https://lmsysorg-dsv4-1.mintlify.site/cookbook/specbundle/supported_models.md)
- [SpecBundle Usage](https://lmsysorg-dsv4-1.mintlify.site/cookbook/specbundle/specbundle_usage.md)
- [Autoregressive Model Benchmark Documentation](https://lmsysorg-dsv4-1.mintlify.site/cookbook/base/benchmarks/autoregressive_model_benchmark.md)
- [Diffusion Models Benchmark Documentation](https://lmsysorg-dsv4-1.mintlify.site/cookbook/base/benchmarks/diffusion_model_benchmark.md)
- [SGLang Diffusion](https://lmsysorg-dsv4-1.mintlify.site/docs/sglang-diffusion/index.md): Accelerated image and video generation with diffusion models.
- [Install SGLang Diffusion](https://lmsysorg-dsv4-1.mintlify.site/docs/sglang-diffusion/installation.md): Install SGLang Diffusion on NVIDIA, AMD, MUSA, and Ascend platforms.
- [Supported Models](https://lmsysorg-dsv4-1.mintlify.site/docs/sglang-diffusion/compatibility_matrix.md): Browse model families and public checkpoints supported by SGLang Diffusion.
- [Disaggregated Diffusion Pipeline](https://lmsysorg-dsv4-1.mintlify.site/docs/sglang-diffusion/disaggregation.md)
- [CLI reference](https://lmsysorg-dsv4-1.mintlify.site/docs/sglang-diffusion/api/cli.md): Run one-off generation tasks and launch the HTTP server from the command line.
- [OpenAI API](https://lmsysorg-dsv4-1.mintlify.site/docs/sglang-diffusion/api/openai_api.md): Image and video generation endpoints with LoRA adapter management.
- [Realtime and Causal Video Models](https://lmsysorg-dsv4-1.mintlify.site/docs/sglang-diffusion/realtime_models.md)
- [Diffusion models with autoregressive stages](https://lmsysorg-dsv4-1.mintlify.site/docs/sglang-diffusion/models_with_ar.md): Run diffusion pipelines with in-process or separately deployed autoregressive encoders.
- [Diffusion Models with Prompt Enhancement (PE)](https://lmsysorg-dsv4-1.mintlify.site/docs/sglang-diffusion/models_with_pe.md): Run ERNIE-Image with built-in prompt enhancement or a separate SGLang-served PE model.
- [Post-Processing](https://lmsysorg-dsv4-1.mintlify.site/docs/sglang-diffusion/api/post_processing.md)
- [Performance Optimization](https://lmsysorg-dsv4-1.mintlify.site/docs/sglang-diffusion/performance-optimization.md): Choose performance levers for SGLang Diffusion by latency, throughput, memory, and quality tradeoffs.
- [Deployment and Performance Modes](https://lmsysorg-dsv4-1.mintlify.site/docs/sglang-diffusion/deployment_cookbook.md): Choose component residency, FSDP, CFG parallelism, SP, TP, and performance-mode presets in SGLang Diffusion.
- [Attention Backends](https://lmsysorg-dsv4-1.mintlify.site/docs/sglang-diffusion/attention_backends.md): Select and configure attention backends for SGLang diffusion pipelines.
- [Fused Kernels](https://lmsysorg-dsv4-1.mintlify.site/docs/sglang-diffusion/fused_kernels.md): The fused CUDA/Triton kernels SGLang Diffusion ships, what each one replaces, and which are on by default.
- [Parallelism Overview](https://lmsysorg-dsv4-1.mintlify.site/docs/sglang-diffusion/parallelism.md)
- [Sequence Parallelism](https://lmsysorg-dsv4-1.mintlify.site/docs/sglang-diffusion/ring_sp_performance.md)
- [Encoder Parallelism](https://lmsysorg-dsv4-1.mintlify.site/docs/sglang-diffusion/encoder_parallel.md)
- [Inference Batching](https://lmsysorg-dsv4-1.mintlify.site/docs/sglang-diffusion/dynamic_batching.md): Batch compatible native SGLang-Diffusion requests during serving.
- [Caching Acceleration](https://lmsysorg-dsv4-1.mintlify.site/docs/sglang-diffusion/caching-acceleration.md): Compare caching acceleration strategies for diffusion models.
- [Cache-DiT Acceleration](https://lmsysorg-dsv4-1.mintlify.site/docs/sglang-diffusion/cache_dit.md): Configure Cache-DiT acceleration for diffusion inference.
- [TeaCache Acceleration](https://lmsysorg-dsv4-1.mintlify.site/docs/sglang-diffusion/teacache.md): Configure TeaCache for temporal similarity-based diffusion acceleration.
- [Spectrum Acceleration](https://lmsysorg-dsv4-1.mintlify.site/docs/sglang-diffusion/spectrum.md): Approximate request-scoped denoising-step acceleration.
- [Progressive Resolution Generation](https://lmsysorg-dsv4-1.mintlify.site/docs/sglang-diffusion/progressive_resolution.md): Experimental spectral progressive resolution growing for selected SGLang Diffusion pipelines.
- [Quantization](https://lmsysorg-dsv4-1.mintlify.site/docs/sglang-diffusion/quantization.md)
- [Profiling](https://lmsysorg-dsv4-1.mintlify.site/docs/sglang-diffusion/profiling.md): Profile SGLang diffusion workloads with PyTorch Profiler and Nsight Systems.
- [Environment Variables](https://lmsysorg-dsv4-1.mintlify.site/docs/sglang-diffusion/environment_variables.md): Configure SGLang diffusion behavior with environment variables.
- [Support New Diffusion Models](https://lmsysorg-dsv4-1.mintlify.site/docs/sglang-diffusion/support_new_models.md): A concise implementation guide for adding diffusion model families to SGLang-Diffusion.
- [CI Performance Baselines](https://lmsysorg-dsv4-1.mintlify.site/docs/sglang-diffusion/ci_perf.md): Generate and update diffusion performance baselines used in CI.
- [Contributing to SGLang Diffusion](https://lmsysorg-dsv4-1.mintlify.site/docs/sglang-diffusion/contributing.md)