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

# Create model endpoint

> Creates a new model gateway endpoint that proxies requests to an external LLM provider. Requires an integration connection with valid API credentials for the target provider.



## OpenAPI

````yaml /api-reference/controlplane.yml post /models
openapi: 3.0.3
info:
  title: Blaxel Control Plane
  version: 0.0.1
servers:
  - description: Blaxel Control Plane
    url: https://api.blaxel.ai/v0
  - description: Blaxel Inference
    url: https://run.blaxel.ai
security:
  - OAuth2:
      - admin
  - ApiKeyAuth: []
paths:
  /models:
    post:
      tags:
        - models
      summary: Create model endpoint
      description: >-
        Creates a new model gateway endpoint that proxies requests to an
        external LLM provider. Requires an integration connection with valid API
        credentials for the target provider.
      operationId: CreateModel
      requestBody:
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/Model'
        required: true
      responses:
        '200':
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Model'
          description: successful operation
        '400':
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Error'
          description: >-
            Bad request - Invalid model configuration or integration connection
            not found
        '401':
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Error'
          description: Unauthorized - Invalid or missing authentication credentials
        '403':
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Error'
          description: Forbidden - Insufficient permissions to create models
        '409':
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Error'
          description: Conflict - Model with this name already exists
        '500':
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Error'
          description: Internal server error
      security:
        - OAuth2:
            - models:create
        - ApiKeyAuth: []
components:
  schemas:
    Model:
      type: object
      description: >-
        Gateway endpoint to external LLM provider APIs (OpenAI, Anthropic, etc.)
        with unified access control, credentials management, and usage tracking.
      properties:
        events:
          $ref: '#/components/schemas/CoreEvents'
        metadata:
          $ref: '#/components/schemas/Metadata'
        spec:
          $ref: '#/components/schemas/ModelSpec'
        status:
          $ref: '#/components/schemas/Status'
      required:
        - metadata
        - spec
    Error:
      type: object
      description: Standard error response returned by the API when a request fails
      properties:
        code:
          type: integer
          description: HTTP status code of the error
          example: 409
        error:
          type: string
          description: Error type or code identifying the kind of error
          example: Resource already exists
        message:
          type: string
          description: Human-readable error message describing what went wrong
          example: Invalid request body
      required:
        - error
    CoreEvents:
      type: array
      description: Events happening on a resource deployed on Blaxel
      items:
        $ref: '#/components/schemas/CoreEvent'
      readOnly: true
    Metadata:
      type: object
      description: >-
        Common metadata fields shared by all Blaxel resources including name,
        labels, timestamps, and ownership information
      allOf:
        - $ref: '#/components/schemas/TimeFields'
        - $ref: '#/components/schemas/OwnerFields'
        - properties:
            displayName:
              type: string
              description: >-
                Human-readable name for display in the UI. Can contain spaces
                and special characters, max 63 characters.
              example: My Resource
            labels:
              $ref: '#/components/schemas/MetadataLabels'
            name:
              type: string
              description: >-
                Unique identifier for the resource within the workspace. Must be
                lowercase alphanumeric with hyphens, max 49 characters.
                Immutable after creation.
              example: my-resource
            plan:
              type: string
              description: >-
                Billing plan tier applied to this resource (inherited from
                workspace account)
              readOnly: true
            url:
              type: string
              description: >-
                Auto-generated endpoint URL for accessing this resource (for
                agents, functions, models, sandboxes)
              readOnly: true
            workspace:
              type: string
              description: >-
                Name of the workspace this resource belongs to (read-only, set
                automatically)
              readOnly: true
          required:
            - name
    ModelSpec:
      type: object
      description: >-
        Configuration for a model gateway endpoint including provider type,
        credentials, and access policies
      properties:
        enabled:
          type: boolean
          description: >-
            When false, the model endpoint is disabled and will not accept
            inference requests
          example: true
          default: true
        flavors:
          $ref: '#/components/schemas/Flavors'
        integrationConnections:
          $ref: '#/components/schemas/IntegrationConnectionsList'
        policies:
          $ref: '#/components/schemas/PoliciesList'
        runtime:
          $ref: '#/components/schemas/ModelRuntime'
        sandbox:
          type: boolean
          description: >-
            When true, uses sandbox/test credentials from the integration
            connection
          example: false
    Status:
      type: string
      description: Deployment status of a resource deployed on Blaxel
      enum:
        - DELETING
        - TERMINATED
        - FAILED
        - DEACTIVATED
        - DEACTIVATING
        - UPLOADING
        - BUILDING
        - DEPLOYING
        - DEPLOYED
        - BUILT
      readOnly: true
    CoreEvent:
      type: object
      description: Core event
      properties:
        canaryRevision:
          type: string
          description: Canary revisionID link to the event
        message:
          type: string
          description: Event message
          example: Deployment successful
        revision:
          type: string
          description: RevisionID link to the event
          example: rev-abc123
        status:
          type: string
          description: Event status
          example: DEPLOYED
        time:
          type: string
          description: Event time
          example: '2025-01-15T10:30:00Z'
        type:
          type: string
          description: Event type
          example: deployment
      readOnly: true
    TimeFields:
      type: object
      description: Time fields for Persistance
      properties:
        createdAt:
          type: string
          description: The date and time when the resource was created
          readOnly: true
        updatedAt:
          type: string
          description: The date and time when the resource was updated
          readOnly: true
    OwnerFields:
      type: object
      description: Owner fields for Persistance
      properties:
        createdBy:
          type: string
          description: The user or service account who created the resource
          readOnly: true
        updatedBy:
          type: string
          description: The user or service account who updated the resource
          readOnly: true
    MetadataLabels:
      type: object
      description: >-
        Key-value pairs for organizing and filtering resources. Labels can be
        used to categorize resources by environment, project, team, or any
        custom taxonomy.
      additionalProperties:
        type: string
    Flavors:
      type: array
      description: Types of hardware available for deployments
      items:
        $ref: '#/components/schemas/Flavor'
    IntegrationConnectionsList:
      type: array
      items:
        type: string
        description: Deployment integration connection name
    PoliciesList:
      type: array
      items:
        type: string
        description: Policy name
    ModelRuntime:
      type: object
      description: >-
        Configuration identifying which external LLM provider and model this
        gateway endpoint proxies to
      properties:
        endpointName:
          type: string
          description: >-
            Provider-specific endpoint name (e.g., HuggingFace Inference
            Endpoints name)
        generation:
          type: string
          description: >-
            Infrastructure generation. Empty (default) uses the classic
            deployment path. mk3 deploys through the model-gateway on microVM
            clusters.
          enum:
            - mk3
            - mk2
        model:
          type: string
          description: >-
            Model identifier at the provider (e.g., gpt-4.1,
            claude-sonnet-4-20250514, mistral-large-latest)
          example: gpt-4.1
        organization:
          type: string
          description: >-
            Organization or account identifier at the provider (required for
            some providers like OpenAI)
          example: org-abc123
        type:
          type: string
          description: >-
            LLM provider type determining the API protocol and authentication
            method
          enum:
            - hf_private_endpoint
            - hf_public_endpoint
            - huggingface
            - public_model
            - mcp
            - openai
            - anthropic
            - gemini
            - mistral
            - deepseek
            - cohere
            - cerebras
            - xai
            - vertexai
            - azure-openai-service
            - azure-ai-inference
            - azure-marketplace
            - groq
          example: openai
    Flavor:
      type: object
      description: A type of hardware available for deployments
      properties:
        name:
          type: string
          description: Flavor name (e.g. t4)
          example: t4
        type:
          type: string
          description: Flavor type (e.g. cpu, gpu)
          enum:
            - cpu
            - gpu
          example: cpu
  securitySchemes:
    OAuth2:
      description: OAuth2 authentication with JWT tokens
      flows:
        authorizationCode:
          authorizationUrl: /oauth/authorize
          scopes:
            admin: Administrative access
            agents:create: Create agents
            agents:delete: Delete agents
            agents:get: Get agent details
            agents:list: List agents
            agents:update: Update agents
            apiKey:list: List API keys
            apiKey:write: Create and delete API keys
            configurations:list: List configurations
            customdomains:create: Create custom domains
            customdomains:delete: Delete custom domains
            customdomains:get: Get custom domain details
            customdomains:list: List custom domains
            customdomains:update: Update custom domains
            functions:create: Create functions
            functions:delete: Delete functions
            functions:get: Get function details
            functions:list: List functions
            functions:update: Update functions
            integrations:create: Create integrations
            integrations:list: List integrations
            invitations:list: List invitations
            jobs:create: Create jobs
            jobs:delete: Delete jobs
            jobs:get: Get job details
            jobs:list: List jobs
            jobs:update: Update jobs
            locations:list: List locations
            mcphub:list: List MCP hub resources
            models:create: Create models
            models:delete: Delete models
            models:get: Get model details
            models:list: List models
            models:update: Update models
            policies:create: Create policies
            policies:delete: Delete policies
            policies:get: Get policy details
            policies:list: List policies
            policies:update: Update policies
            sandboxes:control: Control sandbox operations
            sandboxes:create: Create sandboxes
            sandboxes:delete: Delete sandboxes
            sandboxes:get: Get sandbox details
            sandboxes:list: List sandboxes
            sandboxes:update: Update sandboxes
            sandboxhub:list: List sandbox hub resources
            templates:get: Get template details
            templates:list: List templates
            volumeTemplates:create: Create volume templates
            volumeTemplates:delete: Delete volume templates
            volumeTemplates:get: Get volume template details
            volumeTemplates:list: List volume templates
            volumeTemplates:upsert: Create or update volume templates
            volumes:create: Create volumes
            volumes:delete: Delete volumes
            volumes:get: Get volume details
            volumes:list: List volumes
            workspaces:accept: Accept workspace invitation
            workspaces:create: Create workspaces
            workspaces:decline: Decline workspace invitation
            workspaces:delete: Delete workspaces
            workspaces:get: Get workspace details
            workspaces:invite: Invite users to workspace
            workspaces:leave: Leave workspace
            workspaces:list: List workspaces
            workspaces:remove: Remove users from workspace
            workspaces:update: Update workspaces
          tokenUrl: /oauth/token
        clientCredentials:
          scopes:
            admin: Administrative access
            agents:create: Create agents
            agents:delete: Delete agents
            agents:get: Get agent details
            agents:list: List agents
            agents:update: Update agents
            apiKey:list: List API keys
            apiKey:write: Create and delete API keys
            configurations:list: List configurations
            customdomains:create: Create custom domains
            customdomains:delete: Delete custom domains
            customdomains:get: Get custom domain details
            customdomains:list: List custom domains
            customdomains:update: Update custom domains
            functions:create: Create functions
            functions:delete: Delete functions
            functions:get: Get function details
            functions:list: List functions
            functions:update: Update functions
            integrations:create: Create integrations
            integrations:list: List integrations
            invitations:list: List invitations
            jobs:create: Create jobs
            jobs:delete: Delete jobs
            jobs:get: Get job details
            jobs:list: List jobs
            jobs:update: Update jobs
            locations:list: List locations
            mcphub:list: List MCP hub resources
            models:create: Create models
            models:delete: Delete models
            models:get: Get model details
            models:list: List models
            models:update: Update models
            policies:create: Create policies
            policies:delete: Delete policies
            policies:get: Get policy details
            policies:list: List policies
            policies:update: Update policies
            sandboxes:control: Control sandbox operations
            sandboxes:create: Create sandboxes
            sandboxes:delete: Delete sandboxes
            sandboxes:get: Get sandbox details
            sandboxes:list: List sandboxes
            sandboxes:update: Update sandboxes
            sandboxhub:list: List sandbox hub resources
            templates:get: Get template details
            templates:list: List templates
            volumeTemplates:create: Create volume templates
            volumeTemplates:delete: Delete volume templates
            volumeTemplates:get: Get volume template details
            volumeTemplates:list: List volume templates
            volumeTemplates:upsert: Create or update volume templates
            volumes:create: Create volumes
            volumes:delete: Delete volumes
            volumes:get: Get volume details
            volumes:list: List volumes
            workspaces:accept: Accept workspace invitation
            workspaces:create: Create workspaces
            workspaces:decline: Decline workspace invitation
            workspaces:delete: Delete workspaces
            workspaces:get: Get workspace details
            workspaces:invite: Invite users to workspace
            workspaces:leave: Leave workspace
            workspaces:list: List workspaces
            workspaces:remove: Remove users from workspace
            workspaces:update: Update workspaces
          tokenUrl: /oauth/token
      type: oauth2
    ApiKeyAuth:
      bearerFormat: API Key
      description: Long-lived API key for programmatic access
      scheme: bearer
      type: http

````