vLLM Embeddings API

openapi: 3.1.0
info:
  title: vLLM OpenAI-Compatible Server Audio Embeddings API
  version: '1'
  description: 'vLLM is a high-throughput open-source inference and serving engine for

    LLMs. Running `vllm serve` exposes an OpenAI-compatible REST API plus

    vLLM-specific endpoints. Authentication is via a server-startup

    `--api-key` flag; clients supply it as a Bearer token in the

    Authorization header (matching the OpenAI Python client).

    '
  contact:
    name: vLLM Documentation
    url: https://docs.vllm.ai/en/latest/serving/openai_compatible_server.html
  license:
    name: Apache-2.0
    url: https://www.apache.org/licenses/LICENSE-2.0
servers:
- url: http://{host}:{port}
  description: Local vLLM server
  variables:
    host:
      default: localhost
    port:
      default: '8000'
security:
- bearerAuth: []
tags:
- name: Embeddings
  description: OpenAI-compatible embeddings
paths:
  /v1/embeddings:
    post:
      tags:
      - Embeddings
      summary: Create embeddings
      operationId: createEmbedding
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/EmbeddingRequest'
      responses:
        '200':
          description: Embedding response
components:
  schemas:
    EmbeddingRequest:
      type: object
      required:
      - model
      - input
      properties:
        model:
          type: string
        input:
          oneOf:
          - type: string
          - type: array
            items:
              type: string
  securitySchemes:
    bearerAuth:
      type: http
      scheme: bearer
      description: API key supplied at server startup via --api-key