vLLM

GPU LLM inference server with continuous batching and an OpenAI-compatible API.

Provider: default
Kind: ML inference
Category: Compute
Official docs: https://docs.vllm.ai/en/latest/cli/serve.html

Properties · 4 set by this component

property	type	values	required	kind default	this component
accelerator	enum	none \| gpu \| asic	no	none	▸ gpu
acceleratorModel · advanced	enum	nvidia-h100 \| nvidia-h200 \| nvidia-b200 \| nvidia-l4 \| nvidia-a100 \| amd-mi300x \| aws-trainium \| aws-inferentia \| google-tpu \| intel-gaudi \| groq-lpu \| cerebras-wse	no	—	inherited
apiKeyRequired · component property	boolean	—	no	false	inherited
embeddingSyncMinutes · advanced	integer	—	no	—	inherited
gpuMemoryUtilization · component property	string	—	no	0.92	inherited
inputs	enum	live \| historical \| both	no	live	▸ live
instances	integer	—	no	—	inherited
managed	boolean	—	no	false	▸ false
port · component property	integer	—	no	8000	inherited
snapshotToObjectStore · advanced	boolean	—	no	false	inherited
tensorParallelSize · component property	integer	—	no	1	inherited
training	enum	batch \| online	no	batch	▸ batch

Rules that apply · 1

Every rule whose selector covers this component's kind or its category. Whether a rule fires depends on the design: its properties, its edges and its groups.

Stateful node without zone redundancyerror

stateful-node-without-zone-redundancy

A stateful compute node pinned to exactly one availability zone disappears with that zone, and its state goes with it. Nodes inside a cell are exempt, because there the cell is deliberately single-zone and redundancy comes from the other cells.

— all of these hold:
— the node's `stateless` is false
— the node sits inside exactly 1 availability-zone group
— this does not hold:
— the node sits inside at least one cell group

Use it in a design

nodes:
  - { id: "vllm", component: "vllm" }

Other ML inference products · 37

SageMaker endpointsagemaker-endpointawsml-inference Azure ML endpointazure-ml-endpointazureml-inference Workers AIcloudflare-workers-aicloudflareml-inference Anyscale Endpointsanyscale-endpointsml-inference Bananabanana-devml-inference Basetenbasetenml-inference Beambeam-cloudml-inference BentoMLbentomlml-inference Hugging Face Inference Endpointshuggingface-inferenceml-inference Infinityinfinity-embeddingsml-inference Janjanml-inference KServekserveml-inference llama.cpp serverllama-cpp-serverml-inference llm-dllm-dml-inference LM Studiolm-studioml-inference LMDeploylmdeployml-inference LocalAIlocalaiml-inference Generic ML inferenceml-inferenceml-inference MLC LLMmlc-llmml-inference Modalmodalml-inference Mystic AImystic-aiml-inference NVIDIA NIM (self-hosted)nvidia-nimml-inference OctoAIoctoaiml-inference Ollamaollamaml-inference Ray Serveray-serveml-inference Seldon Coreseldon-coreml-inference SGLangsglangml-inference TensorFlow Servingtensorflow-servingml-inference TensorRT-LLMtensorrt-llmml-inference Text Embeddings Inferencetext-embeddings-inferenceml-inference Text Generation Inferencetext-generation-inferenceml-inference TorchServetorchserveml-inference NVIDIA Tritontritonml-inference Xinferencexinferenceml-inference Vertex AI endpointvertex-ai-endpointgcpml-inference OCI Data Science Model Deploymentoci-data-science-model-deploymentociml-inference OCI Generative AI Dedicated AI Clusteroci-generative-ai-clusterociml-inference

Source

1# yaml-language-server: $schema=../../schema/component.schema.json
2# vLLM: GPU LLM inference server with an OpenAI-compatible HTTP API.
3# Defaults from the "vllm serve" CLI reference: port 8000, no API key set, tensor parallel size 1
4# and gpu-memory-utilization 0.92.
5schemaVersion: "1"
6type: "component"
7id: "vllm"
8kind: "ml-inference"
9provider: "default"
10docs: "https://docs.vllm.ai/en/latest/cli/serve.html"
11defaults:
12  training: "batch"
13  inputs: "live"
14  accelerator: "gpu"
15  managed: false
16properties:
17  port:
18    type: "integer"
19    default: 8000
20  apiKeyRequired:
21    type: "boolean"
22    default: false
23  tensorParallelSize:
24    type: "integer"
25    default: 1
26  gpuMemoryUtilization:
27    type: "string"
28    default: "0.92"

The same page as markdown, for an agent: /catalogue/components/vllm.md