Skip to content
Hayonix

API reference · Schema

DeployableModel

A public catalog model whose weights a dedicated endpoint can serve

  • objectstringrequired

    Always `deployable_model`: the type of this object

    One ofdeployable_model

  • modelstringrequired

    Model ID (`publisher/name`)

  • display_namestringrequired

    Name to show people

  • typestringrequired

    What the model does

    One ofchatcompletionembeddingrerank

  • publisherstring | nullrequired

    Who published the model; null when not recorded

  • context_lengthinteger | nullrequired

    The model's context length in tokens; null when not known

  • weightsobjectrequired

    The weights a deployment of this model serves

    Show 2 fields of weights
    • repositorystringrequired

      The Hugging Face repository the weights come from

    • filestring | nullrequired

      The weights file or quantisation tag used; null when the whole repository is used

  • enginesobjectrequired

    The engine chosen by default on each kind of hardware

    Show 2 fields of engines
    • cpustring | nullrequired

      The engine used on CPU hardware (null: GPU only)

      One ofllama_cppteivllm

    • gpustringrequired

      The engine used on GPU hardware

      One ofllama_cppteivllm

Example
{
  "object": "deployable_model",
  "model": "qwen/qwen2.5-0.5b-instruct",
  "display_name": "string",
  "type": "chat",
  "publisher": "string",
  "context_length": 1,
  "weights": {
    "repository": "string",
    "file": "string"
  },
  "engines": {
    "cpu": "llama_cpp",
    "gpu": "llama_cpp"
  }
}

Used by