Create Reinforcement Fine-tuning Step
Path parameters
The Account Id
Query parameters
ID of the RLOR trainer job, a random UUID will be generated if not specified.
Request body
The name of the dataset used for training.
The name of a separate dataset to use for evaluation.
Whether to auto-carve the dataset for eval.
JobState represents the state an asynchronous job can be in.
- JOB_STATE_PAUSED: Job is paused, typically due to account suspension or manual intervention.
The email address of the user who initiated this fine-tuning job.
A list of reward metrics to use for training in format of "<reward_name>=<weight>".
Rollout deployment name associated with this RLOR trainer job. This is optional. If not set, trainer will not trigger weight sync to rollout engine.
The number of nodes to use for the fine-tuning job. If not specified, the default is 1.
Accelerator seconds used by the job, keyed by accelerator type (e.g., "NVIDIA_H100_80GB"). Updated periodically.
Service-mode RLOR trainers currently support full-parameter tuning only. When enabled, trainingConfig.loraRank must be 0 (loraRank>0 is rejected).
The deployment ID used for hot loading. When set, checkpoints are saved to this deployment's hot load bucket, enabling weight swaps on inference. Only valid for service-mode or keep-alive jobs.
Use dedicated resources for the job. The only supported value currently is "pilot". Defaults to empty.
When true, run the trainer in forward-only mode (no backward/optimizer). Used for reference models in GRPO that only need forward passes.
Response
A successful response.
The name of the dataset used for training.
The name of a separate dataset to use for evaluation.
Whether to auto-carve the dataset for eval.
JobState represents the state an asynchronous job can be in.
- JOB_STATE_PAUSED: Job is paused, typically due to account suspension or manual intervention.
The email address of the user who initiated this fine-tuning job.
A list of reward metrics to use for training in format of "<reward_name>=<weight>".
Rollout deployment name associated with this RLOR trainer job. This is optional. If not set, trainer will not trigger weight sync to rollout engine.
The number of nodes to use for the fine-tuning job. If not specified, the default is 1.
Accelerator seconds used by the job, keyed by accelerator type (e.g., "NVIDIA_H100_80GB"). Updated periodically.
Service-mode RLOR trainers currently support full-parameter tuning only. When enabled, trainingConfig.loraRank must be 0 (loraRank>0 is rejected).
The deployment ID used for hot loading. When set, checkpoints are saved to this deployment's hot load bucket, enabling weight swaps on inference. Only valid for service-mode or keep-alive jobs.
Use dedicated resources for the job. The only supported value currently is "pilot". Defaults to empty.
When true, run the trainer in forward-only mode (no backward/optimizer). Used for reference models in GRPO that only need forward passes.