Эта страница переведена автоматически. Читать английский оригинал.

Component

core/ai-model/vllm

vLLM running in-cluster as a self-hosted inference server. A Recipe synthesizes the runtime directly: a weights-pull initContainer resolves the model pointer into a model-cache volume the serving container mounts, and a ClusterIP Service fronts the OpenAI-compatible API. The model value is a deterministic pointer resolved at pod start, never fetched at compile time. Exports its host, port, endpoint, model, and protocol to consumers via the uses topology, so an agent reaches the model without hardcoding an endpoint. Each realization also publishes where Prometheus finds the server's metrics: one address as `metrics_endpoint`, or, where one address cannot reach every replica, the Service or the pod labels a scraper discovers them by. On KServe the cluster's serving runtime decides the metrics port and path, so nothing is published there. On a cluster running the Prometheus Operator, `core/prometheus-exporter` attached to the synthesized Kubernetes realization, its port set to `http`, emits a ServiceMonitor for it instead.

Происхождение

ОфициальныйПодписано идентичностью подписи, которую этот каталог признаёт официальным издателем.
Дайджест содержимого
sha256:b13caeae34054f971912cfb864d53c9efbe299e59dbd5280ed7ce312e50fed52
Опубликовано
2026-11-10T10:09:41Z
Идентичность подписи
https://token.actions.githubusercontent.com/https://github.com/aggrega-ai/aggrega-hub/.github/workflows/catalog-publish-sign.yml@refs/heads/main

Подпись

  • Официальныйapplication/vnd.dev.sigstore.bundle.v0.3+json

    https://token.actions.githubusercontent.com/https://github.com/aggrega-ai/aggrega-hub/.github/workflows/catalog-publish-sign.yml@refs/heads/main

    sha256:50a6d28f11d3c84b4ec7b3b89bbdecc5d9f1523323fda172c2340fbd633553ce

Ссылка на эту версию

oci://hub.aggrega.ai/component/core/ai-model/vllm:1.0.0

Параметры разработчика

Значения, которые разработчик задаёт в сущности Aggregate.

Расширяет BaseComponents#/schemas/BaseModelParameter

ИмяТипПо умолчаниюОбязательныйОписание
served_model_namestring—нет—

Размеры, задаваемые оператором

Размеры инфраструктуры, которые оператор настраивает для каждого окружения.

ИмяТипПо умолчаниюОбязательныйОписание
portinteger (min 1024, max 65535)8000нетServer listen port for the model API. Most operators should omit this and let the Recipe assign the default (8000).
engine_versionstringv0.6.3нетvLLM runtime image tag. Selects the image tag on the synthesized serving and weights-pull containers.
replicasinteger (min 1)1нетNumber of serving replicas. Each replica holds its own model-cache volume and pulls the weights independently.
min_replicasinteger (min 1)—нетLower bound for the replica count when horizontal autoscaling is configured. Declared ahead of rendering — the synth recipes serve a fixed replica count and do not yet emit an autoscaler; set it to record the intended lower bound.
max_replicasinteger (min 1)—нетUpper bound for the replica count when horizontal autoscaling is configured. Declared ahead of rendering — the synth recipes serve a fixed replica count and do not yet emit an autoscaler; set it to record the intended upper bound.
computeobject—нет—
model_cacheobject—нетPer-replica model-weights cache volume. Sized for the pulled model; each serving replica gets its own claim.
model_artifactstring—нетOptional OCI reference to the model weights, an alternative weights source to the Hugging Face model id in values.model. A deterministic pointer the runtime resolves at pod start; the compiler never fetches it. Declared ahead of rendering — the synth recipes resolve weights from values.model and do not yet consume this pointer.
control_plane_presentbooleanfalseнетOperator affirmation that the model-serving control plane (the KServe or KubeAI controller and its custom-resource definitions) is installed in the target cluster. A controller-driven realization halts the build unless this is set true, so a serving custom resource is never emitted as an orphan into a cluster that cannot reconcile it. The direct synthesis recipes ignore this flag.
overridesobject (open)—нетProvider-specific overrides. Accepts any nested structure. Each Recipe documents which keys it reads.
gpuobject—нетOptional GPU scheduling for the serving container. When set, the Recipe requests GPUs as the extended resource nvidia.com/gpu and, when a runtime class is named, schedules the pod under it. Requires a GPU-capable cluster and, for runtime_class, a matching RuntimeClass configured in the cluster. Omit the whole object to serve on CPU.

Контракт выходных значений

Значения, которые этот компонент публикует для потребителей через топологию uses.

ИмяТипПо умолчаниюОбязательныйОписание
hoststring—даIn-cluster DNS name of the Service fronting the inference server. Which Service that is follows the realization the Manifest selected, so a consumer reads this rather than deriving a name.
portinteger—даTCP port that Service accepts model requests on. A controller-driven realization fronts the server on its own port, so a consumer reads this rather than assuming a default.
endpointstring—даFull base URL of the OpenAI-compatible API the server serves, with the host, the port and the API path already composed. A consumer hands it to a client library as its base URL.
modelstring—даName a request must ask this server for. It is the served name the realization registered, which is not always the weights id the Aggregate declared.
protocolstring—нетURL scheme to reach `host` and `port` over, for a consumer composing its own address rather than reading `endpoint`.
metrics_endpointstring—нетThe full address Prometheus scrapes this server's metrics at — scheme, host, port and path, `http://<host>:<port>/metrics`, on the port the API is served on. Empty where one address cannot reach every replica, which then publishes the service or pod facts instead, and on KServe, which publishes none.
metrics_servicestring—нетName of the Service every replica's metrics sit behind, for a scraper that discovers each endpoint of that Service rather than scraping one address. Set by the Kubernetes synthesis while infra.replicas is above one, beside `metrics_service_port` and `metrics_path`; empty otherwise.
metrics_service_portstring—нетName of the port of `metrics_service` that carries the metrics, `http`. Set and empty together with `metrics_service`.
metrics_pod_selectorstring—нетKubernetes label selector matching every pod that serves this server's metrics, for a scraper that discovers those pods directly because no Service of the realization carries them. Set by the KubeAI realization, beside `metrics_pod_port` and `metrics_path`; empty otherwise.
metrics_pod_portstring—нетName of the container port on those pods that carries the metrics, `http`. Set and empty together with `metrics_pod_selector`.
metrics_pathstring—нетPath the metrics are served at, `/metrics`, published beside the service or the pod facts. Empty where the server publishes `metrics_endpoint`, which carries its own path.

Сокеты uses

Сокеты зависимостей, через которые подключается этот компонент. Цель каждого соединения, проведённого через сокет, должна удовлетворять его селектору меток, и каждое соединение может нести подключение в слотах, которые объявляет сокет. Обязательный слот должен заполнить каждое соединение, которое сущность Aggregate или сущность Release проводит через сокет.

Сокеты uses не объявлены.

Слоты подключения

Именованные точки привязки, в которых Feature подключается к этому компоненту. Селектор слота ограничивает, какая сущность Feature может его заполнить, а его список внедрения указывает, как могут доставляться материалы сущности Feature. Обязательный слот должен быть привязан в каждом экземпляре — подключением в самом экземпляре или в области выше него, — а конфиденциальный слот объявляется для работы с секретными материалами. Feature также может подключаться под именем, которое не объявляет ни один слот.

СлотСелекторОбязательныйВнедрениеКонфиденциальныйОписание
model_pull_tokenfeature.aggrega.ai/kind=secretнетenvдаOptional model-registry pull credential for gated model weights. When attached, the weights-pull initContainer receives it as an environment variable; unattached, weights are pulled anonymously.
api_keyfeature.aggrega.ai/kind=secretнетenvдаOptional key the server requires of every caller. When attached, the Recipe delivers it as VLLM_API_KEY, the variable vLLM reads at startup, and each request under /v1 must then carry it as a bearer token in the Authorization header. The server checks this one key, so every caller must present the same key. Unattached, the server requires none.

Список разрешённых самовнедрений

env

Связи

Типизированные рёбра к другим опубликованным сущностям и от них. Ссылка без опубликованной цели отображается как разорванное ребро, а не как ссылка.

Использует

Используется в

Справочник по понятию Component →

Каталог Aggrega — открытый указатель компонентов, функций, рецептов и плагинов, которые их рендерят.

Создано на общей дизайн-системе Aggrega.

Конфиденциальность