- **[How OME works](architecture/how-ome-works.md)** How the OME manager, the model agent and OME's custom resources turn a model and a runtime into serving pods. - **[Deployment modes and OMENative](architecture/deployment-modes.md)** OME runs InferenceService components as Kubernetes Deployments or on OMENative, which creates, updates, repairs and moves each replica as one unit. Opt in to OMENative, see how OME picks a component's mode, and read what OMENative creates and reports.

- **[OMENative overview](omenative/overview.md)** OMENative runs each replica of an InferenceService component as an Instance, one pod or a leader and its workers, that OME creates, updates, repairs, moves and removes as one unit. Multi-node engines and decoders use it by default, and any component can opt in. - **[Instances](omenative/instances.md)** An Instance is one replica of an OMENative component, a single pod or a leader and its workers, whose identity, phase, operation and failures OME records on the component's InferenceReplica. - **[OMENative update strategies](architecture/omenative-update-strategies.md)** Choose how an OMENative rollout replaces an Instance's pods, by starting new ones first, recreating them or updating them in place, and pace it with partition, maxSurge and maxUnavailable. - **[Instance restart policy](omenative/instance-restart-policy.md)** Choose whether OMENative repairs only a failed pod or rebuilds its whole Instance, and see how a restart runs. - **[Migration and transient scale](omenative/migration-and-transient-scale.md)** OMENative moves an Instance to another node by starting its replacement first, and can run more or fewer Instances for a while without an edit to the InferenceService.

- **[Gang scheduling](serving/gang-scheduling.md)** With the scheduler-plugins PodGroup CRD installed, OME creates a PodGroup for each multi-pod OMENative Instance, so a gang-aware scheduler places its leader and workers together or not at all. - **[The OME scheduler](scheduling/ome-scheduler.md)** An alpha second scheduler that places each gang of pods in one accelerator domain, and binds its pods only when the whole gang fits. - **[Alfred](scheduling/alfred.md)** Alfred, OME's alpha GPU cluster caretaker, recommends Instance moves when free GPUs fragment or a node turns unhealthy and, when you let it, asks OMENative to carry them out. - **[Alfred policies](scheduling/alfred-policies.md)** How Alfred's defragmentation and node-health policies find Instances to move, and how its arbiter and safety bounds decide which become recommendations or migration requests.

- **[Base models](models/base-models.md)** BaseModel and ClusterBaseModel tell OME where a model's weights live and which nodes get them; OME parses the model's architecture, size and capabilities. - **[Fine-tuned weights](models/fine-tuned-weights.md)** Serve weights fine-tuned from a base model, such as a LoRA adapter, from a zip file in OCI Object Storage.

- **[Serving runtimes](runtimes/serving-runtimes.md)** A ServingRuntime or ClusterServingRuntime describes how to run a model server: its pods, and optionally the models it supports, so OME can choose it for an InferenceService. - **[Runtime inheritance](runtimes/runtime-inheritance.md)** Write one runtime per engine, and keep what differs for each model or GPU in small child runtimes that inherit from it. - **[Runtime revisions and pinning](runtimes/runtime-revisions.md)** Pin an InferenceService to a snapshot of its runtime, choose when runtime changes reach it, and roll back to an earlier snapshot. - **[Accelerator classes](runtimes/accelerator-classes.md)** An AcceleratorClass names one type of GPU and the nodes that have it, so an InferenceService can run its engine and decoder on that type, picked by name or by policy.

- **[InferenceService](serving/inference-services.md)** An InferenceService names a model, a serving runtime or both, and declares the engine, decoder and router components that OME turns into workloads. - **[Component autoscaling](serving/component-autoscaling.md)** Each InferenceService component can carry its own autoscaler block, which picks HPA, KEDA, External or None and the metrics or triggers to scale on. - **[Autoscaler policy](serving/autoscaler-policy.md)** An AutoscalerPolicy is an alpha, reusable set of KEDA or HPA autoscaling settings that InferenceService components share by name. - **[Benchmarks](serving/benchmarks.md)** A BenchmarkJob runs genai-bench against an InferenceService with the traffic scenarios you choose, and stores the results.

- **[Rollout groups](rollouts-and-traffic/rollout-groups.md)** Rollout groups roll an InferenceService's OMENative components out as one, so prefill and decode change together, by canary, blue-green or rolling update, with metric-gated canary steps. - **[Rollout policy](rollouts-and-traffic/rollout-policy.md)** A RolloutPolicy is an alpha, reusable canary, blue-green or rolling-update progression that InferenceService rollout groups attach by reference. - **[Ingress and external access](rollouts-and-traffic/ingress.md)** How clients outside the cluster reach an InferenceService: the Service that OME creates by default, or the Kubernetes Ingress or Gateway API HTTPRoutes it creates when you turn on ingress. - **[Traffic policy](rollouts-and-traffic/traffic-policy.md)** Set an InferenceService's load balancing, session affinity and endpoint override in spec.traffic. OME writes them into the gateway's backend policy and reports the result in status. - **[Traffic map](rollouts-and-traffic/traffic-map.md)** A TrafficMap is the alpha routing table that OME writes for a multi-cluster InferenceService: each workload cluster's share of the requests, and the reasons behind it.