Sign in
RL

Create model resources

Provisions a standalone model resource that training sessions can attach to.

post/rl/model-resources

Request body

base_modelstring required

Base model to provision the resource for

lora_enabledboolean

Whether the resource hosts LoRA sessions or a single full-weight session

Example request

{
  "base_model": "Qwen/Qwen3-0.6B",
  "compute_config": {
    "num_generator_replicas": 2,
    "gpu_type": "B200-SXM"
  },
  "lora_enabled": true
}

Response

Model resource details

idstring required

Unique identifier for the model resource

status'MODEL_RESOURCES_STATUS_PENDING' | 'MODEL_RESOURCES_STATUS_CREATING' | 'MODEL_RESOURCES_STATUS_READY' | 'MODEL_RESOURCES_STATUS_ERROR' | 'MODEL_RESOURCES_STATUS_STOPPED' | 'MODEL_RESOURCES_STATUS_STOPPING' required

Lifecycle status of a model resource

base_modelstring required

Base model the resource is provisioned for

lora_enabledboolean required

Whether the resource hosts LoRA sessions or a full-weight session

created_atstring date-time required

Timestamp when the model resource was created

updated_atstring date-time required

Timestamp when the model resource was last updated

created_bystring required

ID of the user who created the model resource

Example response

{
  "id": "123e4567-e89b-12d3-a456-426614174000",
  "error": {
    "message": "Timed out waiting for cluster capacity. Try again later.",
    "occurred_at": "2026-01-02T00:00:05Z"
  },
  "base_model": "Qwen/Qwen3-0.6B",
  "compute_config": {
    "num_generator_replicas": 1,
    "gpu_type": "H100-80GB"
  },
  "lora_enabled": true,
  "created_at": "2026-01-02T00:00:00Z",
  "updated_at": "2026-01-02T00:00:05Z",
  "created_by": "user_123"
}

Changes

Changed in 3 of the 15 revisions of this API.16

  • f4885bc384242See the full diff
    • ○

      added the new optional request property compute_config/gpu_type

      new-optional-request-property

    • ○

      added the optional property compute_config/gpu_type to the response with the 200 status

      response-optional-property-added

  • 42c9ead8f2f511See the full diff
    • ●

      added the new MODEL_RESOURCES_ERROR_CODE_CAPACITY_WAIT_TIMEOUT enum value to the error/code response property for the response status 200

      response-property-enum-value-added

    • ○

      removed the MODEL_RESOURCES_ERROR_CODE_CAPACITY_UNAVAILABLE enum value from the error/code response property for the response status 200

      response-property-enum-value-removed

  • 0c4fdad774353See the full diff
    • ○

      added the new optional request property optimizer_config

      new-optional-request-property

    • ○

      added the required property created_by to the response with the 200 status

      response-required-property-added

    • ○

      added the required property optimizer_config to the response with the 200 status

      response-required-property-added

    This revision also has 2 changes that name no endpoint, such as unreferenced schemas being removed. See the revision's changelog