# tinker.types.TrainingRun

## _class_ [**tinker.types.TrainingRun**](https://github.com/thinking-machines-lab/tinker/blob/main/src/tinker/types/training_run.py#L9)( _[BaseModel](https://github.com/thinking-machines-lab/tinker/blob/main/src/tinker/_models.py#L88)_)

**Fields:**

- [**training\_run\_id**](https://github.com/thinking-machines-lab/tinker/blob/main/src/tinker/types/training_run.py#L10) ( _str_) – The unique identifier for the training run
- [**base\_model**](https://github.com/thinking-machines-lab/tinker/blob/main/src/tinker/types/training_run.py#L13) ( _str_) – The base model name this model is derived from
- [**model\_owner**](https://github.com/thinking-machines-lab/tinker/blob/main/src/tinker/types/training_run.py#L16) ( _str_) – The owner/creator of this model
- [**is\_lora**](https://github.com/thinking-machines-lab/tinker/blob/main/src/tinker/types/training_run.py#L19) ( _bool_) – Whether this model uses LoRA (Low-Rank Adaptation)
- [**corrupted**](https://github.com/thinking-machines-lab/tinker/blob/main/src/tinker/types/training_run.py#L22) ( _bool_) – Whether the model is in a corrupted state
- [**lora\_rank**](https://github.com/thinking-machines-lab/tinker/blob/main/src/tinker/types/training_run.py#L25) ( _int \| None_) – The LoRA rank if this is a LoRA model, null otherwise
- [**last\_request\_time**](https://github.com/thinking-machines-lab/tinker/blob/main/src/tinker/types/training_run.py#L28) ( _datetime_) – The timestamp of the last request made to this model
- [**last\_checkpoint**](https://github.com/thinking-machines-lab/tinker/blob/main/src/tinker/types/training_run.py#L31) ( _[Checkpoint](https://tinker-docs.thinkingmachines.ai/tinker/api-reference/types/checkpoint/) \| None_) – The most recent training checkpoint, if available
- [**last\_sampler\_checkpoint**](https://github.com/thinking-machines-lab/tinker/blob/main/src/tinker/types/training_run.py#L34) ( _[Checkpoint](https://tinker-docs.thinkingmachines.ai/tinker/api-reference/types/checkpoint/) \| None_) – The most recent sampler checkpoint, if available
- [**user\_metadata**](https://github.com/thinking-machines-lab/tinker/blob/main/src/tinker/types/training_run.py#L37) ( _dict[str, str] \| None_) – Optional metadata about this training run, set by the end-user

## Referenced by

- [tinker.types.TrainingRunsResponse](https://tinker-docs.thinkingmachines.ai/tinker/api-reference/types/trainingrunsresponse/)