{ "@context": { "@version": 1.1, "@vocab": "https://schema.org/", "infer": "https://raw.githubusercontent.com/api-evangelist/scalable-inference-serving/main/json-ld/scalable-inference-serving-context.jsonld#", "kserve": "https://kserve.github.io/ns/", "oip": "https://github.com/kserve/open-inference-protocol/ns/", "InferenceRequest": { "@id": "infer:InferenceRequest", "@type": "@id", "comment": "An OIP V2 inference request containing input tensors submitted to a model serving endpoint." }, "InferenceResponse": { "@id": "infer:InferenceResponse", "@type": "@id", "comment": "The response from an OIP V2 inference request containing output tensors." }, "ModelMetadata": { "@id": "infer:ModelMetadata", "@type": "@id", "comment": "Metadata describing a deployed model's inputs, outputs, and platform." }, "InferenceServer": { "@id": "infer:InferenceServer", "@type": "@id", "comment": "An OIP-compliant server (KServe, Triton, BentoML) hosting model inference endpoints." }, "Tensor": { "@id": "infer:Tensor", "@type": "@id", "comment": "A multi-dimensional array of data representing a model input or output." }, "ModelRegistry": { "@id": "infer:ModelRegistry", "@type": "@id", "comment": "A versioned store of machine learning models and their metadata (e.g., MLflow Model Registry)." }, "InferenceService": { "@id": "kserve:InferenceService", "@type": "@id", "comment": "A KServe Custom Resource that deploys a model serving endpoint on Kubernetes." }, "id": "@id", "model_name": "schema:name", "model_version": "schema:version", "platform": "infer:platform", "inputs": { "@id": "infer:inputs", "@container": "@list" }, "outputs": { "@id": "infer:outputs", "@container": "@list" }, "name": "schema:name", "shape": "infer:shape", "datatype": "infer:datatype", "data": "infer:data", "parameters": "infer:parameters", "live": "infer:live", "ready": "infer:ready", "versions": { "@id": "schema:version", "@container": "@list" }, "extensions": { "@id": "infer:extensions", "@container": "@list" }, "tags": "schema:keywords" } }