Skip to main content
POST
/
v1
/
accounts
/
{account_id}
/
rlorTrainerJobs
Create Reinforcement Fine-tuning Step
curl --request POST \
  --url https://api.fireworks.ai/v1/accounts/{account_id}/rlorTrainerJobs \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "displayName": "<string>",
  "dataset": "<string>",
  "evaluationDataset": "<string>",
  "evalAutoCarveout": true,
  "trainingConfig": {
    "outputModel": "<string>",
    "baseModel": "<string>",
    "warmStartFrom": "<string>",
    "jinjaTemplate": "<string>",
    "learningRate": 123,
    "maxContextLength": 123,
    "loraRank": 123,
    "region": "REGION_UNSPECIFIED",
    "epochs": 123,
    "batchSize": 123,
    "gradientAccumulationSteps": 123,
    "learningRateWarmupSteps": 123
  },
  "rewardWeights": [
    "<string>"
  ],
  "wandbConfig": {
    "enabled": true,
    "apiKey": "<string>",
    "project": "<string>",
    "entity": "<string>",
    "runId": "<string>"
  },
  "keepAlive": true,
  "rolloutDeploymentName": "<string>",
  "lossConfig": {
    "method": "METHOD_UNSPECIFIED",
    "klBeta": 123
  }
}
'
{
  "name": "<string>",
  "displayName": "<string>",
  "createTime": "2023-11-07T05:31:56Z",
  "completedTime": "2023-11-07T05:31:56Z",
  "dataset": "<string>",
  "evaluationDataset": "<string>",
  "evalAutoCarveout": true,
  "state": "JOB_STATE_UNSPECIFIED",
  "status": {
    "code": "OK",
    "message": "<string>"
  },
  "createdBy": "<string>",
  "trainingConfig": {
    "outputModel": "<string>",
    "baseModel": "<string>",
    "warmStartFrom": "<string>",
    "jinjaTemplate": "<string>",
    "learningRate": 123,
    "maxContextLength": 123,
    "loraRank": 123,
    "region": "REGION_UNSPECIFIED",
    "epochs": 123,
    "batchSize": 123,
    "gradientAccumulationSteps": 123,
    "learningRateWarmupSteps": 123
  },
  "rewardWeights": [
    "<string>"
  ],
  "wandbConfig": {
    "enabled": true,
    "apiKey": "<string>",
    "project": "<string>",
    "entity": "<string>",
    "runId": "<string>",
    "url": "<string>"
  },
  "keepAlive": true,
  "rolloutDeploymentName": "<string>",
  "lossConfig": {
    "method": "METHOD_UNSPECIFIED",
    "klBeta": 123
  }
}

Authorizations

​
Authorization
string
header
required

Bearer authentication using your Fireworks API key. Format: Bearer <API_KEY>

Path Parameters

​
account_id
string
required

The Account Id

Query Parameters

​
rlorTrainerJobId
string

ID of the RLOR trainer job, a random UUID will be generated if not specified.

Body

application/json
​
displayName
string
​
dataset
string

The name of the dataset used for training.

​
evaluationDataset
string

The name of a separate dataset to use for evaluation.

​
evalAutoCarveout
boolean

Whether to auto-carve the dataset for eval.

​
trainingConfig
BaseTrainingConfig contains common configuration fields shared across different training job types. Next ID: 19 · object

Common training configurations.

​
rewardWeights
string[]

A list of reward metrics to use for training in format of "<reward_name>=".

​
wandbConfig
object

The Weights & Biases team/user account for logging training progress.

​
keepAlive
boolean
​
rolloutDeploymentName
string

Rollout deployment name associated with this RLOR trainer job. This is optional. If not set, trainer will not trigger weight sync to rollout engine.

​
lossConfig
object

Reinforcement learning loss method + hyperparameters for the underlying trainer.

Response

200 - application/json

A successful response.

​
name
string
​
displayName
string
​
createTime
string<date-time>
​
completedTime
string<date-time>
​
dataset
string

The name of the dataset used for training.

​
evaluationDataset
string

The name of a separate dataset to use for evaluation.

​
evalAutoCarveout
boolean

Whether to auto-carve the dataset for eval.

​
state
enum<string>
default:JOB_STATE_UNSPECIFIED

JobState represents the state an asynchronous job can be in.

  • JOB_STATE_PAUSED: Job is paused, typically due to account suspension or manual intervention.
Available options:
JOB_STATE_UNSPECIFIED,
JOB_STATE_CREATING,
JOB_STATE_RUNNING,
JOB_STATE_COMPLETED,
JOB_STATE_FAILED,
JOB_STATE_CANCELLED,
JOB_STATE_DELETING,
JOB_STATE_WRITING_RESULTS,
JOB_STATE_VALIDATING,
JOB_STATE_DELETING_CLEANING_UP,
JOB_STATE_PENDING,
JOB_STATE_EXPIRED,
JOB_STATE_RE_QUEUEING,
JOB_STATE_CREATING_INPUT_DATASET,
JOB_STATE_IDLE,
JOB_STATE_CANCELLING,
JOB_STATE_EARLY_STOPPED,
JOB_STATE_PAUSED
​
status
Mimics [https://github.com/googleapis/googleapis/blob/master/google/rpc/status.proto] · object
​
createdBy
string

The email address of the user who initiated this fine-tuning job.

​
trainingConfig
BaseTrainingConfig contains common configuration fields shared across different training job types. Next ID: 19 · object

Common training configurations.

​
rewardWeights
string[]

A list of reward metrics to use for training in format of "<reward_name>=".

​
wandbConfig
object

The Weights & Biases team/user account for logging training progress.

​
keepAlive
boolean
​
rolloutDeploymentName
string

Rollout deployment name associated with this RLOR trainer job. This is optional. If not set, trainer will not trigger weight sync to rollout engine.

​
lossConfig
object

Reinforcement learning loss method + hyperparameters for the underlying trainer.