TrainingGpuCapacityItem

class baseten.client.managementapi.TrainingGpuCapacityItem(*, gpu_type, baseline, limit, usage_count, dedicated_usage_count=0, spot_usage_count=0)

Bases: BaseModel

Parameters:

• gpu_type (Annotated[str, FieldInfo(annotation=NoneType, required=True, title='Gpu Type', description='GPU type identifier (e.g. H100, A100-40GB)')])

• baseline (Annotated[int, FieldInfo(annotation=NoneType, required=True, title='Baseline', description='Baseline GPU allocation; jobs below this threshold are expected to run immediately. 0 if not configured.')])

• limit (Annotated[int, FieldInfo(annotation=NoneType, required=True, title='Limit', description='Maximum concurrent GPUs of this type for this org')])

• usage_count (Annotated[int, FieldInfo(annotation=NoneType, required=True, title='Usage Count', description='GPUs currently in use by active training jobs')])

• dedicated_usage_count (Annotated[int | None, FieldInfo(annotation=NoneType, required=True, title='Dedicated Usage Count', description='Portion of usage_count from dedicated (on-demand) jobs, which run against the baseline.')])

• spot_usage_count (Annotated[int | None, FieldInfo(annotation=NoneType, required=True, title='Spot Usage Count', description='Portion of usage_count from spot jobs, which burst into the peak and may push usage above the limit.')])

model_config = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].