Ta strona została przetłumaczona maszynowo. Przeczytaj oryginał w języku angielskim.

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.

Pochodzenie

OficjalnyPodpisano tożsamością podpisującą, którą ten katalog uznaje za oficjalnego wydawcę.
Skrót zawartości
sha256:b13caeae34054f971912cfb864d53c9efbe299e59dbd5280ed7ce312e50fed52
Opublikowano
2026-11-10T10:09:41Z
Tożsamość podpisująca
https://token.actions.githubusercontent.com/https://github.com/aggrega-ai/aggrega-hub/.github/workflows/catalog-publish-sign.yml@refs/heads/main

Podpis

  • Oficjalnyapplication/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

Referencja w tej wersji

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

Parametry dewelopera

Wartości ustawiane przez dewelopera w encji Aggregate.

Rozszerza BaseComponents#/schemas/BaseModelParameter

NazwaTypWartość domyślnaWymaganeOpis
served_model_namestring—nie—

Wymiarowanie przez operatora

Wymiarowanie infrastruktury, które operator dostraja dla każdego środowiska.

NazwaTypWartość domyślnaWymaganeOpis
portinteger (min 1024, max 65535)8000nieServer listen port for the model API. Most operators should omit this and let the Recipe assign the default (8000).
engine_versionstringv0.6.3nievLLM runtime image tag. Selects the image tag on the synthesized serving and weights-pull containers.
replicasinteger (min 1)1nieNumber of serving replicas. Each replica holds its own model-cache volume and pulls the weights independently.
min_replicasinteger (min 1)—nieLower 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)—nieUpper 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—nie—
model_cacheobject—niePer-replica model-weights cache volume. Sized for the pulled model; each serving replica gets its own claim.
model_artifactstring—nieOptional 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_presentbooleanfalsenieOperator 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)—nieProvider-specific overrides. Accepts any nested structure. Each Recipe documents which keys it reads.
gpuobject—nieOptional 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.

Kontrakt wyjściowy

Wartości, które ten komponent publikuje konsumentom za pośrednictwem topologii uses.

NazwaTypWartość domyślnaWymaganeOpis
hoststring—takIn-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—takTCP 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—takFull 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—takName 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—nieURL scheme to reach `host` and `port` over, for a consumer composing its own address rather than reading `endpoint`.
metrics_endpointstring—nieThe 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—nieName 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—nieName of the port of `metrics_service` that carries the metrics, `http`. Set and empty together with `metrics_service`.
metrics_pod_selectorstring—nieKubernetes 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—nieName of the container port on those pods that carries the metrics, `http`. Set and empty together with `metrics_pod_selector`.
metrics_pathstring—niePath 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.

Gniazda uses

Gniazda zależności, przez które łączy się ten komponent. Cel każdego połączenia poprowadzonego przez gniazdo musi spełniać jego selektor etykiet, a każde połączenie może przenosić dołączenie w slotach deklarowanych przez gniazdo. Wymagany slot musi wypełnić każde połączenie, które encja Aggregate lub encja Release prowadzi przez gniazdo.

Nie zadeklarowano gniazd uses.

Sloty dołączania

Nazwane punkty wiązania, w których encja Feature dołącza się do tego komponentu. Selektor slotu ogranicza, która encja Feature może go wypełnić, a jego lista inject określa, jak można dostarczyć materiał encji Feature. Wymagany slot musi być powiązany w każdej instancji, przez dołączenie w instancji lub w zakresie nad nią, a slot wrażliwy jest zadeklarowany do obsługi materiału tajnego. Encja Feature może również dołączyć się pod nazwą, której nie deklaruje żaden slot.

SlotSelektorWymaganeWstrzykiwanieWrażliweOpis
model_pull_tokenfeature.aggrega.ai/kind=secretnieenvtakOptional 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=secretnieenvtakOptional 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.

Lista dozwolonych samowstrzyknięć

env

Relacje

Typowane krawędzie do innych opublikowanych encji i od nich. Odwołanie bez opublikowanego celu jest renderowane jako przerwana krawędź, a nie jako łącze.

Używa

Używane przez

Dokumentacja pojęcia Component →

Katalog Aggrega — otwarty indeks komponentów, funkcji, receptur oraz wtyczek, które je renderują.

Zbudowano przy użyciu wspólnego systemu projektowego Aggrega.

Prywatność