Scalable Inference Serving · Capability

KServe Open Inference Protocol API — Inference

KServe Open Inference Protocol API — Inference. 2 operations. Lead operation: Run Model Inference. Self-contained Naftiko capability covering one Scalable Inference Serving business surface.

Run with Naftiko Scalable Inference ServingInference

What You Can Do

POST
Runinference — Run Model Inference
/v1/v2/models/{model-name}/infer
POST
Runmodelversioninference — Run Model Version Inference
/v1/v2/models/{model-name}/versions/{model-version}/infer

MCP Tools

run-model-inference

Run Model Inference

run-model-version-inference

Run Model Version Inference

Capability Spec

kserve-open-inference-protocol-inference.yaml Raw ↑ 
naftiko: 1.0.0-alpha2
info:
  label: KServe Open Inference Protocol API — Inference
  description: 'KServe Open Inference Protocol API — Inference. 2 operations. Lead operation: Run Model Inference. Self-contained
    Naftiko capability covering one Scalable Inference Serving business surface.'
  tags:
  - Scalable Inference Serving
  - Inference
  created: '2026-05-19'
  modified: '2026-05-19'
binds:
- namespace: env
  keys:
    SCALABLE_INFERENCE_SERVING_API_KEY: SCALABLE_INFERENCE_SERVING_API_KEY
capability:
  consumes:
  - type: http
    namespace: kserve-open-inference-protocol-inference
    baseUri: https://inference.kserve.example.com
    description: KServe Open Inference Protocol API — Inference business capability. Self-contained, no shared references.
    resources:
    - name: v2-models-model_name-infer
      path: /v2/models/{model_name}/infer
      operations:
      - name: runinference
        method: POST
        description: Run Model Inference
        outputRawFormat: json
        outputParameters:
        - name: result
          type: object
          value: $.
        inputParameters:
        - name: model_name
          in: path
          type: string
          description: Name of the model to run inference against.
          required: true
        - name: body
          in: body
          type: object
          description: Request body (JSON).
          required: true
    - name: v2-models-model_name-versions-model_version-infer
      path: /v2/models/{model_name}/versions/{model_version}/infer
      operations:
      - name: runmodelversioninference
        method: POST
        description: Run Model Version Inference
        outputRawFormat: json
        outputParameters:
        - name: result
          type: object
          value: $.
        inputParameters:
        - name: model_name
          in: path
          type: string
          required: true
        - name: model_version
          in: path
          type: string
          required: true
        - name: body
          in: body
          type: object
          description: Request body (JSON).
          required: true
  exposes:
  - type: rest
    namespace: kserve-open-inference-protocol-inference-rest
    port: 8080
    description: REST adapter for KServe Open Inference Protocol API — Inference. One Spectral-compliant resource per consumed
      operation, prefixed with /v1.
    resources:
    - path: /v1/v2/models/{model-name}/infer
      name: v2-models-model-name-infer
      description: REST surface for v2-models-model_name-infer.
      operations:
      - method: POST
        name: runinference
        description: Run Model Inference
        call: kserve-open-inference-protocol-inference.runinference
        with:
          model_name: rest.model_name
          body: rest.body
        outputParameters:
        - type: object
          mapping: $.
    - path: /v1/v2/models/{model-name}/versions/{model-version}/infer
      name: v2-models-model-name-versions-model-version-infer
      description: REST surface for v2-models-model_name-versions-model_version-infer.
      operations:
      - method: POST
        name: runmodelversioninference
        description: Run Model Version Inference
        call: kserve-open-inference-protocol-inference.runmodelversioninference
        with:
          model_name: rest.model_name
          model_version: rest.model_version
          body: rest.body
        outputParameters:
        - type: object
          mapping: $.
  - type: mcp
    namespace: kserve-open-inference-protocol-inference-mcp
    port: 9090
    transport: http
    description: MCP adapter for KServe Open Inference Protocol API — Inference. One tool per consumed operation, routed inline
      through this capability's consumes block.
    tools:
    - name: run-model-inference
      description: Run Model Inference
      hints:
        readOnly: false
        destructive: false
        idempotent: false
      call: kserve-open-inference-protocol-inference.runinference
      with:
        model_name: tools.model_name
        body: tools.body
      outputParameters:
      - type: object
        mapping: $.
    - name: run-model-version-inference
      description: Run Model Version Inference
      hints:
        readOnly: false
        destructive: false
        idempotent: false
      call: kserve-open-inference-protocol-inference.runmodelversioninference
      with:
        model_name: tools.model_name
        model_version: tools.model_version
        body: tools.body
      outputParameters:
      - type: object
        mapping: $.