> ## Documentation Index
> Fetch the complete documentation index at: https://docs.omnia-voice.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Create dedicated endpoint

> Provision a new dedicated GPU endpoint. Requires an owner/admin key and enough prepaid balance for the minimum runway. Discover valid model/GPU/region/count combinations from the templates endpoint first.



## OpenAPI

````yaml /openapi.json post /v1/dedicated
openapi: 3.1.0
info:
  title: Omnia Management API
  description: >-
    The management API behind the improvement loop: capture and setup, request
    logs and datasets, grades (labels), judges (criteria), evals and deploy
    gates, fine-tuning and reinforcement learning, dedicated GPU endpoints, and
    model aliases and versions. Authenticated with a workspace API key
    (sk_sovereign_...). The inference API (chat, embeddings, rerank, responses)
    is OpenAI-compatible and documented separately.


    Responses are snake_case, list endpoints on the loop products use the
    {"object": "list", "data": [...]} envelope, and refusals use the same nested
    error shape the gateway emits: {"error": {"message", "type", "code"}}.
    Request bodies on the loop products (logs, labels, criteria, evals,
    datasets, aliases) are snake_case; the training and infrastructure products
    (fine-tuning, GRPO, environment tools, dedicated, model-version adoption)
    validate camelCase bodies, and each schema below says which it is. Endpoints
    that spend money require a key minted by a workspace owner or admin and
    return 403 otherwise.
  version: 1.0.0
servers:
  - url: https://gateway.omnia-voice.com
    description: Production
  - url: https://platform.omnia-voice.com/api
    description: Production (legacy alias — same API, older base URL)
security:
  - bearerAuth: []
paths:
  /v1/dedicated:
    post:
      tags:
        - Dedicated endpoints
      summary: Create dedicated endpoint
      description: >-
        Provision a new dedicated GPU endpoint. Requires an owner/admin key and
        enough prepaid balance for the minimum runway. Discover valid
        model/GPU/region/count combinations from the templates endpoint first.
      operationId: createDedicatedEndpoint
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/CreateDedicatedEndpoint'
      responses:
        '201':
          description: The created endpoint.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/DedicatedEndpoint'
        '400':
          $ref: '#/components/responses/BadRequest'
        '401':
          $ref: '#/components/responses/Unauthorized'
        '402':
          $ref: '#/components/responses/InsufficientBalance'
        '403':
          $ref: '#/components/responses/Forbidden'
components:
  schemas:
    CreateDedicatedEndpoint:
      type: object
      description: >-
        This body is camelCase, unlike the snake_case loop endpoints. The
        service validates the model/flavor/GPU/region combo against the deploy
        templates, prices it, and requires prepaid balance for the first hour.
      required:
        - name
        - modelName
        - flavorName
        - gpuType
        - gpuCount
        - region
        - minReplicas
        - maxReplicas
      properties:
        name:
          type: string
        modelName:
          type: string
          description: A template name from GET /v1/dedicated/templates.
        flavorName:
          type: string
          description: A flavor of that template.
        gpuType:
          type: string
        gpuCount:
          type: integer
        region:
          type: string
        minReplicas:
          type: integer
          description: At least 1.
        maxReplicas:
          type: integer
          description: At least minReplicas.
        description:
          type: string
        customWeightsId:
          type: string
          description: >-
            Serve a fine-tuned model's weights by artifact id (used when
            deploying a trained model).
        fineTuningJobId:
          type: string
          description: >-
            The source fine-tuning job id, when deploying a model trained on
            this platform.
    DedicatedEndpoint:
      type: object
      properties:
        id:
          type: string
        name:
          type: string
        description:
          type:
            - string
            - 'null'
        model_name:
          type: string
        flavor_name:
          type: string
        gpu_type:
          type: string
        gpu_count:
          type: integer
        region:
          type: string
        min_replicas:
          type: integer
        max_replicas:
          type: integer
        status:
          type: string
          description: Live-reconciled from the control plane on every list/get.
          enum:
            - STARTING
            - UPDATING
            - RUNNING
            - STOPPING
            - STOPPED
            - DELETED
            - WARNING
        enabled:
          type: boolean
          description: >-
            Whether the endpoint is started. Billing runs only while RUNNING and
            enabled.
        hourly_rate_usd:
          type: number
          description: >-
            Your price per GPU-hour for this configuration (rate frozen at
            deploy).
        markup_bps:
          type: integer
        pending_cost_usd:
          type: number
          description: Cost accrued since the last metering sweep, at the frozen rate.
        routing_key:
          type:
            - string
            - 'null'
          description: >-
            Call the endpoint through the gateway as model
            "dedicated/<routing_key>".
        base_url:
          type: string
          description: The inference base URL to point your client at.
        last_metered_at:
          type: string
        created_at:
          type: string
    Error:
      type: object
      properties:
        error:
          type: object
          properties:
            message:
              type: string
            type:
              type: string
              description: >-
                invalid_request_error, insufficient_quota, rate_limit_error, or
                api_error.
            code:
              type: string
              description: >-
                Machine-stable cause, e.g. invalid_api_key, not_found,
                insufficient_permissions, precondition_failed.
          required:
            - message
            - type
            - code
      description: >-
        Every refusal — gateway and management API alike — uses this one
        envelope.
  responses:
    BadRequest:
      description: Malformed request or invalid field.
      content:
        application/json:
          schema:
            $ref: '#/components/schemas/Error'
    Unauthorized:
      description: Missing, malformed, or revoked API key.
      content:
        application/json:
          schema:
            $ref: '#/components/schemas/Error'
          example:
            error:
              message: Invalid API key
              type: invalid_request_error
              code: invalid_api_key
    InsufficientBalance:
      description: Your wallet can't cover the required prepaid runway.
      content:
        application/json:
          schema:
            $ref: '#/components/schemas/Error'
          example:
            error:
              message: >-
                Insufficient balance: deploying this endpoint requires at least
                1h of runway. Top up and try again.
              type: insufficient_quota
              code: insufficient_balance
    Forbidden:
      description: The key lacks the required owner/admin permission.
      content:
        application/json:
          schema:
            $ref: '#/components/schemas/Error'
  securitySchemes:
    bearerAuth:
      type: http
      scheme: bearer
      description: >-
        Your workspace API key, e.g. `sk_sovereign_...`, sent as `Authorization:
        Bearer <key>`.

````