> For clean Markdown content of this page, append .md to this URL. For the complete documentation index, see https://docs.nvidia.com/dynamo/llms.txt. For full content including API reference and SDK examples, see https://docs.nvidia.com/dynamo/llms-full.txt.

# User Guides

## Docs

- [Disaggregated Serving](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/disaggregated-serving.md): Find optimal prefill/decode configuration for disaggregated serving deployments
- [Router Guide](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/kv-cache-aware-routing.md): Deployment modes, quick start, and page map for Dynamo routing docs
- [KVBM Guide](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/kv-cache-offloading.md): Enable KV offloading using KV Block Manager (KVBM) for Dynamo deployments
- [Multimodal Model Serving](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/multimodal.md): Deploy multimodal models with image, video, and audio support in Dynamo
- [Embedding Cache](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/multimodal/embedding-cache.md): Cache vision encoder embeddings to skip re-encoding repeated images
- [Encoder Disaggregation](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/multimodal/encoder-disaggregation.md): Separate vision encoding into a dedicated worker for independent scaling
- [Multimodal KV Routing](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/multimodal/multimodal-kv-routing.md): Route multimodal requests to workers with the best KV cache overlap
- [Diffusion](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/diffusion.md): Deploy diffusion models for text-to-image, text-to-video, and more in Dynamo
- [FastVideo](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/diffusion/fastvideo.md)
- [vLLM-Omni](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/diffusion/v-llm-omni.md)
- [Diffusion](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/diffusion/sg-lang-diffusion.md)
- [Video Diffusion Support (Experimental)](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/diffusion/trt-llm-diffusion.md)
- [LoRA Adapters](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/lo-ra-adapters.md): Serve fine-tuned LoRA adapters with dynamic loading and routing in Dynamo
- [Agents](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/agents.md): Workload-aware inference with agentic hints for routing, scheduling, and KV cache Management
- [Agent Context and Tracing](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/agents/agent-context-and-tracing.md): Attach workflow identity to agentic requests
- [SGLang for Agentic Workloads](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/agents/sg-lang-for-agentic-workloads.md): Priority scheduling and session control for multi-turn agentic serving
- [Chat Processor Options](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/chat-processors.md): Choose the right preprocessing pipeline for tool calling, reasoning, and tokenization
- [Tool Calling](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/tool-calling.md): Connect Dynamo to external tools and services using function calling
- [Reasoning](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/reasoning.md): Configure reasoning parsers for models that emit thinking content
- [Observability (Local)](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/observability-local.md): Monitor Dynamo deployments with metrics, logging, and tracing
- [Prometheus + Grafana Setup](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/observability-local/prometheus-grafana-setup.md)
- [Metrics](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/observability-local/metrics.md)
- [Metrics Developer Guide](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/observability-local/metrics-developer-guide.md)
- [Health Checks](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/observability-local/health-checks.md)
- [Tracing](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/observability-local/tracing.md)
- [Logging](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/observability-local/logging.md)
- [Fault Tolerance](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/fault-tolerance.md): Handle failures gracefully with request migration, cancellation, and graceful shutdown
- [Request Migration](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/fault-tolerance/request-migration.md)
- [Request Cancellation](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/fault-tolerance/request-cancellation.md)
- [Request Rejection](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/fault-tolerance/request-rejection.md)
- [Graceful Shutdown](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/fault-tolerance/graceful-shutdown.md)
- [Testing](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/fault-tolerance/testing.md)
- [Dynamo Benchmarking](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/benchmarking.md): Benchmark and compare performance across Dynamo deployment configurations
- [Mocker](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/mocker/testing-with-mocker.md)
- [Mocker Trace Replay](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/mocker/mocker-offline-trace-replay.md): Replay Mooncake-style traces through the mocker in offline or online mode
- [Writing Python Workers in Dynamo](https://docs.nvidia.com/dynamo/v1.1.0/user-guides/writing-python-workers.md): Create custom Python workers and engines for Dynamo