RL

Review an RL setup without reserving compute

Changed on

Authenticates the caller and prepares a server-owned recipe. Preview does not create a workload or reserve compute. An unavailable estimate is not a zero cost estimate. Launchability remains subject to current billing and capacity checks at submission and provisioning.

post/v1/rl-runs/preview

Request

  • Base URL
    / (relative, and the document was published on a file host, not at the API’s origin)
  • Auth
    HTTP bearer

Request body

recipe_id'reasoning-gym-leg-counting'
recipe_version'1.0.0'
mode'evaluate' | 'train'
evaluation_tasksinteger
training_stepsinteger
max_cost_usdnumber

Deprecated compatibility field. Customer spending limits are not enforced.

seedinteger
include_tracesboolean

Opt in to prompt, model response and verifier text in live task events.

Response

Success

review_tokenstring required

Empty when the runtime is unavailable. Otherwise submit unchanged with the reviewed configuration.

launchableboolean required
blocking_reasonsstring[] required

Changes

    • breaking

      the response property // became optional for the status

    • info

      the request property default value 5 was removed

    • info

      request property deprecated

    • info

      the response's property default value 5 was removed for the status

    • info

      the response's property default value 5 was removed for the status

    • info

      response property configuration/max_cost_usd deprecated

    • info

      response property recipe/defaults/max_cost_usd deprecated

    • warning

      the request property's max was set to 100.00

    • warning

      the request property's min was set to 1.00

    • info

      the request property default value 5 was added

    • info

      request property reactivated

    • info

      the response property // became required for the status

    • info

      the response's property default value 5 was added for the status

    • info

      the response's property default value 5 was added for the status

    • info

      response property configuration/max_cost_usd reactivated

    • info

      response property recipe/defaults/max_cost_usd reactivated