v1
ApplicationAPI & CLI
Double column
API
CLI
JS Client

Llm Model Deployments /

Create LLM model deployment

Create an LLM model deployment

Required permission

Project > AiModels > General > Create

Path parameters

    • projectId

      string required
      ID of the project

Request body

  • {object}

    • name

      string required
      The name of the LLM model deployment.
      min length
      3
      max length
      100
      pattern
      ^[a-zA-Z0-9]+((-|\s)[a-zA-Z0-9]+)*$

    • description

      string
      A description of the LLM model deployment.
      max length
      1024
      pattern
      ^[\s\S]*$

    • tags

      [array]
      An array of previously defined tags to help identify and group the resource.
      • string
        min length
        3
        max length
        100
        pattern
        ^[a-zA-Z0-9]+(-[a-zA-Z0-9]+)*$

    • ports

      [array]
      • {object}

        • name

          string required
          The name used to identify the port.
          min length
          1
          max length
          8
          pattern
          ^[a-zA-Z](-?[a-zA-Z0-9]+(-[a-zA-Z0-9]+)*)?$

        • internalPort

          integer required
          The port number.
          min
          1
          max
          65535

        • public

          boolean
          If true, the port will be exposed publicly.

        • domains

          [array]
          An array of domains to redirect to this port. Each domain must first be verified and registered to your account.
          • string
            A domain to redirect to this port.

        • disableNfDomain

          boolean
          Disable routing on the default code.run domain for public HTTP ports with custom domains.

        • protocol

          string required
          The protocol to use for the port.
          one of
          HTTP, HTTP/2, TCP, UDP

    • spec

      {object} required
API
CLI
JS Client

POST /v1/projects/{projectId}/llm-model-deployments

Example request

Request body
curl
curl --header "Content-Type: application/json" \
  --header "Authorization: Bearer NORTHFLANK_API_TOKEN" \
  --request POST \
  --data '{"name":"deepseek-v3-deployment","description":"A deployment for DeepSeek V3 model","spec":{"configuration":{"nodeConfiguration":{"storage":{"storageClass":"nvme","storageSize":6144}}}}}' \
  https://api.northflank.com/v1/projects/{projectId}/llm-model-deployments

Example response

200 OK

success

© 2026 Northflank Ltd. All rights reserved.

northflank.comTermsPrivacyfeedback@northflank.com