RL

List supported models

Returns the models supported by the RL service and their limits for training/sampling operations.

get/rl/supported-models

Response

List of supported RL models

Example response

{
  "data": [
    {
      "base_model": "Qwen/Qwen3-0.6B",
      "trainer_config": {
        "full": {
          "max_batch_size": 512,
          "max_seq_length": 40960
        },
        "lora": {
          "max_batch_size": 512,
          "max_seq_length": 40960,
          "max_rank": 64
        }
      },
      "generator_config": {
        "context_length": 40960,
        "sampling_defaults": {
          "n": 1,
          "max_tokens": 512,
          "temperature": 1,
          "logprobs": 1
        }
      }
    }
  ]
}

Changes

Changed in 1 of the 14 revisions of this API.11

  • 0c4fdad7743511See the full diff
    • the data/items/generator_config/sampling_defaults/temperature response's property type/format changed from number/float to number/ for status 200

      response-property-type-changed

    • added the non-success response with the status 429

      response-non-success-status-added

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