generated: '2026-07-21'
method: derived
status: candidate
source: openapi/rightnow-openapi.json
note: No official hosted/remote RunInfra MCP server was found (docs, GitHub org, MCP registries). This
  is a DERIVED candidate tool list mapped from the OpenAI-compatible REST operations, offered as a proposal
  — not a published server.
api: RunInfra Inference API
base_url: https://api.runinfra.ai/v1
auth:
  type: bearer
  header: Authorization
tools:
- name: create_chat_completion
  operationId: createChatCompletion
  method: POST
  path: /chat/completions
  description: OpenAI-compatible chat completion with streaming, tools, structured output.
- name: create_response
  operationId: createResponse
  method: POST
  path: /responses
  description: Responses-shaped adapter over chat-completions deployments.
- name: create_embedding
  operationId: createEmbedding
  method: POST
  path: /embeddings
  description: Vector embeddings for semantic search, RAG, and clustering.
- name: rerank_documents
  operationId: createRerank
  method: POST
  path: /rerank
  description: Text and multimodal document reranking.
- name: create_image
  operationId: createImage
  method: POST
  path: /images/generations
  description: Text-to-image generation.
- name: create_transcription
  operationId: createTranscription
  method: POST
  path: /audio/transcriptions
  description: Speech-to-text transcription (multipart upload).
- name: create_speech
  operationId: createSpeech
  method: POST
  path: /audio/speech
  description: Text-to-speech synthesis (binary response).
- name: list_models
  operationId: listModels
  method: GET
  path: /models
  description: List verified deployed models in the workspace.
- name: retrieve_model
  operationId: retrieveModel
  method: GET
  path: /models/{model}
  description: Retrieve a single deployed model by id.
deployment:
  mode: none
  verified: derived
  tools: 9
  checked: '2026-08-12'
  source: catalog MCP census