# Tensormesh Operator Documentation

- [Getting Started](https://docs.tensormesh.ai/installation/getting-started.md): Pick an install method for the Tensormesh Operator and verify your cluster is ready.
- [Compatibility Matrix / Supported Models](https://docs.tensormesh.ai/installation/compatibility.md): Models, serving engines, cache features, and recovery scenarios supported on the latest Tensormesh release — and the exact versions to deploy.
- [Install with Helm](https://docs.tensormesh.ai/installation/helm.md): Install the Tensormesh Operator from the official Helm chart.
- [E2E Quickstart](https://docs.tensormesh.ai/installation/example.md): Install the operator, point vLLM at the engine, run an inference, and verify the KV cache is reused.
- [Modify an Existing vLLM Deployment](https://docs.tensormesh.ai/installation/existing-deployment.md): Patch an existing vLLM deployment to consume the LMCache engine created by the Tensormesh Operator.
- [Troubleshooting](https://docs.tensormesh.ai/installation/troubleshooting.md): Diagnose common install and day-2 issues with the Tensormesh Operator.
- [Non-Prefix KV Caching](https://docs.tensormesh.ai/configuration/cacheblend.md): Reuse KV cache beyond prefix boundaries with CacheBlend
- [External Storage Offloading](https://docs.tensormesh.ai/configuration/filesystem-offloading.md): Use a filesystem-backed L2 cache with the Tensormesh Operator by mounting storage into the engine pod and configuring the raw L2 adapter.
- [KVCache Encryption](https://docs.tensormesh.ai/configuration/kvcache-encryption.md): Encrypt KV cache bytes at rest in the L2 tier with aesgcm, using a master key you provide as a Kubernetes Secret.
- [Peer-to-Peer KV Transfer](https://docs.tensormesh.ai/configuration/p2p.md): Enable direct KV cache transfer between engine pods so warm prefixes can be read from any peer in the fleet, not just the local L1 or a shared L2 backend.
- [PD Disaggregation](https://docs.tensormesh.ai/configuration/pd-disaggregation.md): Separate the prefill and decode phases onto dedicated vLLM instances to improve TTFT and GPU utilization at scale.
- [NVIDIA Dynamo](https://docs.tensormesh.ai/integrations/dynamo.md): Offload KV cache from Dynamo vLLM workers to the Tensormesh engine over cross-pod CUDA IPC.
- [Multi-Tenancy](https://docs.tensormesh.ai/management/multi-tenancy.md): Isolate cache reuse and capacity across tenants, users, and environments with cache_salt and per-tenant quotas.
- [Observability](https://docs.tensormesh.ai/observability.md): Export LMCache engine metrics and traces via an OpenTelemetry Collector — to in-cluster Prometheus/Tempo or an external backend.
- [Metrics](https://docs.tensormesh.ai/observability/metrics.md): How LMCache engine metrics are exposed by the operator, which Prometheus metrics matter first, and where to go deeper.
- [CLI](https://docs.tensormesh.ai/reference/cli.md): tmo-cli — the Tensormesh operator/admin CLI for observing and operating LMCacheEngine fleets on Kubernetes.
- [Overview](https://docs.tensormesh.ai/ui/introduction.md): The Tensormesh Operator UI — a live dashboard for your LMCache fleets: cache health, capacity, and GPU visibility in one place.
- [Installation](https://docs.tensormesh.ai/ui/installation.md): Install the Operator UI into your cluster, next to the Tensormesh Operator.
- [Configuration](https://docs.tensormesh.ai/ui/configuration.md): Connect metric history, GPU telemetry, and additional clusters — everything else works out of the box.
- [Access & Security](https://docs.tensormesh.ai/ui/access-and-security.md): Who can reach the dashboard, how to put SSO in front of it, and why it's safe to run in production.
- [Monitoring Your Fleet](https://docs.tensormesh.ai/ui/monitoring-your-fleet.md): Overview, Fleet Map, Metrics, Health, and Thresholds — what each screen shows and when to use it.
- [Thresholds & Notifications](https://docs.tensormesh.ai/ui/thresholds-and-notifications.md): Decide what counts as a problem, and get notified the moment it happens.
- [Troubleshooting & FAQ](https://docs.tensormesh.ai/ui/troubleshooting.md): Common setup problems and short answers to frequent questions.
- [Architecture](https://docs.tensormesh.ai/reference/architecture.md): What the chart deploys, how the pieces connect, and which resources are cluster-scoped (and can orphan).
- [Configuration](https://docs.tensormesh.ai/reference/configuration.md): Every values.yaml key for the Tensormesh Operator chart, plus ready-to-use example overlays.
- [Glossary / Vocabulary](https://docs.tensormesh.ai/reference/glossary.md): Plain-language definitions of the terms used throughout the Tensormesh Operator docs.

## OpenAPI Specs

- [serverless.openapi](/api-reference/serverless.openapi.yaml)
- [common.openapi](/api-reference/src/common.openapi.yaml)
- [activity_service.openapi](/tensormesh-api/specs/activity/v1/activity_service.openapi.json)
- [admin_billing_service.openapi](/tensormesh-api/specs/billing/v1/admin_billing_service.openapi.json)
- [billing_address_service.openapi](/tensormesh-api/specs/billing/v1/billing_address_service.openapi.json)
- [billing_model_service.openapi](/tensormesh-api/specs/billing/v1/billing_model_service.openapi.json)
- [billing_transaction_service.openapi](/tensormesh-api/specs/billing/v1/billing_transaction_service.openapi.json)
- [pricing_service.openapi](/tensormesh-api/specs/billing/v1/pricing_service.openapi.json)
- [product_service.openapi](/tensormesh-api/specs/billing/v1/product_service.openapi.json)
- [stripe_service.openapi](/tensormesh-api/specs/billing/v1/stripe_service.openapi.json)
- [user_balance_service.openapi](/tensormesh-api/specs/billing/v1/user_balance_service.openapi.json)
- [admin_model_service.openapi](/tensormesh-api/specs/model/v1/admin_model_service.openapi.json)
- [model_service.openapi](/tensormesh-api/specs/model/v1/model_service.openapi.json)
- [model.openapi](/tensormesh-api/specs/model/v1/model.openapi.json)
- [observability_service.openapi](/tensormesh-api/specs/observability/v1/observability_service.openapi.json)
- [reserved_deployment_service.openapi](/tensormesh-api/specs/support/v1/reserved_deployment_service.openapi.json)
- [support_service.openapi](/tensormesh-api/specs/support/v1/support_service.openapi.json)
- [admin_user_service.openapi](/tensormesh-api/specs/user/v1/admin_user_service.openapi.json)
- [user_service.openapi](/tensormesh-api/specs/user/v1/user_service.openapi.json)