Sign in

v1

latestOpenAPI 3.1.02026-07-26106201349.6 KB
Gateway

Create Supervised Fine-tuning Job

post/v1/accounts/{account_id}/supervisedFineTuningJobs

Path parameters

account_idstring required

The Account Id

Query parameters

supervisedFineTuningJobIdstring

ID of the supervised fine-tuning job, a random UUID will be generated if not specified.

Request body

namestring
displayNamestring
createTimestring date-time
completedTimestring date-time
datasetstring required

The name of the dataset used for training.

state'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' | 'JOB_STATE_DELETED' | 'JOB_STATE_ARCHIVED'

JobState represents the state an asynchronous job can be in.

  • JOB_STATE_PAUSED: Job is paused, typically due to account suspension or manual intervention.
  • JOB_STATE_DELETED: Job has been deleted.
  • JOB_STATE_ARCHIVED: User-facing state for jobs whose row is retained post-delete (e.g. RLOR trainers within the checkpoint retention window). The internal row is still in JOB_STATE_DELETED; the gateway translates it to ARCHIVED on public responses.
createdBystring

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

outputModelstring

The model ID to be assigned to the resulting fine-tuned model. If not specified, the job ID will be used.

baseModelstring

The name of the base model to be fine-tuned Only one of 'base_model' or 'warm_start_from' should be specified.

warmStartFromstring

The PEFT addon model in Fireworks format to be fine-tuned from Only one of 'base_model' or 'warm_start_from' should be specified.

jinjaTemplatestring

Deprecated: literal Jinja templates are not supported by Training V2. Conversation rendering is selected from the base model's registered renderer configuration instead.

earlyStopboolean

Deprecated: early stopping is not supported by managed training.

epochsinteger

The number of epochs to train for.

learningRatenumber float

The learning rate used for training.

maxContextLengthinteger

The maximum context length to use with the model.

loraRankinteger

The rank of the LoRA layers.

evaluationDatasetstring

The name of a separate dataset to use for evaluation.

isTurboboolean

Whether to run the fine-tuning job in turbo mode.

evalAutoCarveoutboolean

Whether to auto-carve the dataset for eval.

updateTimestring date-time

The update time for the supervised fine-tuning job.

nodesinteger

Deprecated: multi-node scheduling is now handled by the cookbook orchestrator in V2 workflows. This field is ignored for V2 jobs and will be removed in a future release.

batchSizeinteger

Deprecated: legacy V1 token budget. Training V2 batches by samples via batch_size_samples.

mtpEnabledboolean

Deprecated: MTP is no longer supported by managed training. This field is retained for API wire compatibility only.

mtpNumDraftTokensinteger

Deprecated: see mtp_enabled.

mtpFreezeBaseModelboolean

Deprecated: see mtp_enabled.

metricsFileSignedUrlstring
trainerLogsSignedUrlstring

The signed URL for the trainer logs file (stdout/stderr). Only populated if the account has trainer log reading enabled.

renderSamplesSignedUrlstring

The signed URL for orchestrator-rendered token IDs and loss masks.

gradientAccumulationStepsinteger

Deprecated: legacy V1 gradient accumulation. Training V2 batches by samples via batch_size_samples and rejects this field when set.

learningRateWarmupStepsinteger
batchSizeSamplesinteger

The number of samples per gradient batch.

optimizerWeightDecaynumber float

Weight decay (L2 regularization) for optimizer.

purpose'PURPOSE_UNSPECIFIED' | 'PURPOSE_PILOT'

Scheduling purpose for training jobs and deployments.

encryptionState'ENCRYPTION_STATE_UNSPECIFIED' | 'ENCRYPTION_STATE_PLAINTEXT' | 'ENCRYPTION_STATE_CMEK'

EncryptionState is the authoritative, per-resource CMEK marker: the single source of truth for whether a resource's customer-data artifacts are encrypted. It is stamped at resource creation from the owning account's CMEK config and is immutable for the resource's life, so a resource always reads back with the state it was created under regardless of the account's current flag. The encrypt-on-write / decrypt-on-read decision for the resource's data keys off this field.

  • ENCRYPTION_STATE_UNSPECIFIED: Unstamped/legacy rows that predate CMEK. Treated as PLAINTEXT.
  • ENCRYPTION_STATE_PLAINTEXT: Customer-data artifacts are stored as plaintext (Fireworks-owned, no CMEK).
  • ENCRYPTION_STATE_CMEK: Customer-data artifacts are CMEK-encrypted; reads MUST decrypt (fail-closed).

Response

A successful response.

namestring
displayNamestring
createTimestring date-time
completedTimestring date-time
datasetstring required

The name of the dataset used for training.

state'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' | 'JOB_STATE_DELETED' | 'JOB_STATE_ARCHIVED'

JobState represents the state an asynchronous job can be in.

  • JOB_STATE_PAUSED: Job is paused, typically due to account suspension or manual intervention.
  • JOB_STATE_DELETED: Job has been deleted.
  • JOB_STATE_ARCHIVED: User-facing state for jobs whose row is retained post-delete (e.g. RLOR trainers within the checkpoint retention window). The internal row is still in JOB_STATE_DELETED; the gateway translates it to ARCHIVED on public responses.
createdBystring

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

outputModelstring

The model ID to be assigned to the resulting fine-tuned model. If not specified, the job ID will be used.

baseModelstring

The name of the base model to be fine-tuned Only one of 'base_model' or 'warm_start_from' should be specified.

warmStartFromstring

The PEFT addon model in Fireworks format to be fine-tuned from Only one of 'base_model' or 'warm_start_from' should be specified.

jinjaTemplatestring

Deprecated: literal Jinja templates are not supported by Training V2. Conversation rendering is selected from the base model's registered renderer configuration instead.

earlyStopboolean

Deprecated: early stopping is not supported by managed training.

epochsinteger

The number of epochs to train for.

learningRatenumber float

The learning rate used for training.

maxContextLengthinteger

The maximum context length to use with the model.

loraRankinteger

The rank of the LoRA layers.

evaluationDatasetstring

The name of a separate dataset to use for evaluation.

isTurboboolean

Whether to run the fine-tuning job in turbo mode.

evalAutoCarveoutboolean

Whether to auto-carve the dataset for eval.

updateTimestring date-time

The update time for the supervised fine-tuning job.

nodesinteger

Deprecated: multi-node scheduling is now handled by the cookbook orchestrator in V2 workflows. This field is ignored for V2 jobs and will be removed in a future release.

batchSizeinteger

Deprecated: legacy V1 token budget. Training V2 batches by samples via batch_size_samples.

mtpEnabledboolean

Deprecated: MTP is no longer supported by managed training. This field is retained for API wire compatibility only.

mtpNumDraftTokensinteger

Deprecated: see mtp_enabled.

mtpFreezeBaseModelboolean

Deprecated: see mtp_enabled.

metricsFileSignedUrlstring
trainerLogsSignedUrlstring

The signed URL for the trainer logs file (stdout/stderr). Only populated if the account has trainer log reading enabled.

renderSamplesSignedUrlstring

The signed URL for orchestrator-rendered token IDs and loss masks.

gradientAccumulationStepsinteger

Deprecated: legacy V1 gradient accumulation. Training V2 batches by samples via batch_size_samples and rejects this field when set.

learningRateWarmupStepsinteger
batchSizeSamplesinteger

The number of samples per gradient batch.

optimizerWeightDecaynumber float

Weight decay (L2 regularization) for optimizer.

purpose'PURPOSE_UNSPECIFIED' | 'PURPOSE_PILOT'

Scheduling purpose for training jobs and deployments.

encryptionState'ENCRYPTION_STATE_UNSPECIFIED' | 'ENCRYPTION_STATE_PLAINTEXT' | 'ENCRYPTION_STATE_CMEK'

EncryptionState is the authoritative, per-resource CMEK marker: the single source of truth for whether a resource's customer-data artifacts are encrypted. It is stamped at resource creation from the owning account's CMEK config and is immutable for the resource's life, so a resource always reads back with the state it was created under regardless of the account's current flag. The encrypt-on-write / decrypt-on-read decision for the resource's data keys off this field.

  • ENCRYPTION_STATE_UNSPECIFIED: Unstamped/legacy rows that predate CMEK. Treated as PLAINTEXT.
  • ENCRYPTION_STATE_PLAINTEXT: Customer-data artifacts are stored as plaintext (Fireworks-owned, no CMEK).
  • ENCRYPTION_STATE_CMEK: Customer-data artifacts are CMEK-encrypted; reads MUST decrypt (fail-closed).