Fastly AI Accelerator

openapi: 3.0.3
info:
  title: Fastly AI Accelerator
  description: |
    Fastly AI Accelerator is a semantic caching solution that boosts the
    performance of popular LLMs like OpenAI and Google Gemini by 9x. Semantic
    caching maps queries to concepts as vectors so the system can cache answers
    to similar questions regardless of exact wording. AI Accelerator exposes
    drop-in compatible chat completions endpoints that proxy to upstream
    providers while serving cached responses from the edge.
  version: '1.0.0'
servers:
  - url: https://api.fastly.ai
    description: Fastly AI Accelerator edge endpoint
security:
  - FastlyKey: []
tags:
  - name: Chat Completions
  - name: Embeddings
paths:
  /openai/v1/chat/completions:
    post:
      tags: [Chat Completions]
      summary: Create Chat Completion Via OpenAI
      description: OpenAI-compatible chat completion request, semantically cached at the Fastly edge.
      operationId: createOpenAiChatCompletion
      requestBody:
        required: true
        content:
          application/json:
            schema: { $ref: '#/components/schemas/ChatCompletionRequest' }
      responses:
        '200':
          description: Chat completion response
          content:
            application/json:
              schema: { $ref: '#/components/schemas/ChatCompletionResponse' }
  /gemini/v1/models/{model}:generateContent:
    parameters:
      - { in: path, name: model, required: true, schema: { type: string, example: gemini-1.5-pro } }
    post:
      tags: [Chat Completions]
      summary: Generate Content Via Google Gemini
      operationId: generateGeminiContent
      responses:
        '200': { description: Gemini response }
  /openai/v1/embeddings:
    post:
      tags: [Embeddings]
      summary: Create Embeddings
      operationId: createEmbeddings
      responses:
        '200': { description: Embeddings response }
components:
  securitySchemes:
    FastlyKey:
      type: apiKey
      in: header
      name: Fastly-Key
  schemas:
    ChatCompletionRequest:
      type: object
      required: [model, messages]
      properties:
        model: { type: string, example: gpt-4o-mini }
        messages:
          type: array
          items:
            type: object
            properties:
              role: { type: string, enum: [system, user, assistant, tool] }
              content: { type: string }
        temperature: { type: number, format: float }
        max_tokens: { type: integer }
        stream: { type: boolean }
    ChatCompletionResponse:
      type: object
      properties:
        id: { type: string }
        object: { type: string }
        created: { type: integer }
        model: { type: string }
        choices:
          type: array
          items:
            type: object
            properties:
              index: { type: integer }
              message:
                type: object
                properties:
                  role: { type: string }
                  content: { type: string }
              finish_reason: { type: string }
        usage:
          type: object
          properties:
            prompt_tokens: { type: integer }
            completion_tokens: { type: integer }
            total_tokens: { type: integer }
        x_fastly_cache:
          type: object
          description: Fastly semantic cache metadata
          properties:
            status: { type: string, enum: [HIT, MISS, SIMILAR] }
            similarity: { type: number, format: float }