# LoRA

Configuration classes for the pipeline LoRA feature.

## LoRAFeatureConfig

- *class* qairt.experimental.pipeline.torch.llm.lora.configs.LoRAFeatureConfig(*\**, *adapters: dict[str, [qairt.experimental.pipeline.torch.llm.lora.configs.LoRAAdapterConfig](https://docs.qualcomm.com/doc/80-87189-2/topic/qairt-pipeline-lora.html#qairt.experimental.pipeline.torch.llm.lora.configs.LoRAAdapterConfig)] = None*, *use\_cases: dict[str, [qairt.experimental.pipeline.torch.llm.lora.configs.LoRAUseCaseConfig](https://docs.qualcomm.com/doc/80-87189-2/topic/qairt-pipeline-lora.html#qairt.experimental.pipeline.torch.llm.lora.configs.LoRAUseCaseConfig)] = None*, *quant\_updatable\_mode: Literal['adapter\_only', 'all', 'none'] = 'adapter\_only'*)

    - Bases: `BaseModel`

Top-level LoRA feature configuration for the pipeline.

- *field* adapters*: dict[str, [LoRAAdapterConfig](https://docs.qualcomm.com/doc/80-87189-2/topic/qairt-pipeline-lora.html#qairt.experimental.pipeline.torch.llm.lora.configs.LoRAAdapterConfig)]*  *[Optional]*

    - - Validated by

    - - `_check_use_case_adapter_refs`

- model\_computed\_fields*: ClassVar[dict[str, ComputedFieldInfo]]*  *= {}*

    - A dictionary of computed field names and their corresponding <cite>ComputedFieldInfo</cite> objects.

- model\_config*: ClassVar[ConfigDict]*  *= {}*

    - Configuration for the model, should be a dictionary conforming to [<cite>ConfigDict</cite>][pydantic.config.ConfigDict].

- model\_fields*: ClassVar[dict[str, FieldInfo]]*  *= {'adapters': FieldInfo(annotation=dict[str, LoRAAdapterConfig], required=False, default\_factory=dict), 'quant\_updatable\_mode': FieldInfo(annotation=Literal['adapter\_only', 'all', 'none'], required=False, default='adapter\_only'), 'use\_cases': FieldInfo(annotation=dict[str, LoRAUseCaseConfig], required=False, default\_factory=dict)}*

    - Metadata about the fields defined on the model,
mapping of field names to [<cite>FieldInfo</cite>][pydantic.fields.FieldInfo].

This replaces <cite>Model.__fields__</cite> from Pydantic V1.

- *field* quant\_updatable\_mode*: Literal['adapter\_only', 'all', 'none']*  *= 'adapter\_only'*

    - - Validated by

    - - `_check_use_case_adapter_refs`

- *field* use\_cases*: dict[str, [LoRAUseCaseConfig](https://docs.qualcomm.com/doc/80-87189-2/topic/qairt-pipeline-lora.html#qairt.experimental.pipeline.torch.llm.lora.configs.LoRAUseCaseConfig)]*  *[Optional]*

    - - Validated by

    - - `_check_use_case_adapter_refs`

## LoRAAdapterConfig

- *class* qairt.experimental.pipeline.torch.llm.lora.configs.LoRAAdapterConfig(*\**, *path: Path*)

    - Bases: `BaseModel`

Configuration for a single LoRA adapter.

- model\_computed\_fields*: ClassVar[dict[str, ComputedFieldInfo]]*  *= {}*

    - A dictionary of computed field names and their corresponding <cite>ComputedFieldInfo</cite> objects.

- model\_config*: ClassVar[ConfigDict]*  *= {}*

    - Configuration for the model, should be a dictionary conforming to [<cite>ConfigDict</cite>][pydantic.config.ConfigDict].

- model\_fields*: ClassVar[dict[str, FieldInfo]]*  *= {'path': FieldInfo(annotation=Path, required=True, metadata=[PathType(path\_type='dir')])}*

    - Metadata about the fields defined on the model,
mapping of field names to [<cite>FieldInfo</cite>][pydantic.fields.FieldInfo].

This replaces <cite>Model.__fields__</cite> from Pydantic V1.

- *field* path*: DirectoryPath*  *[Required]*

    - - Constraints

    - - **path\_type** = dir

## LoRAUseCaseConfig

- *class* qairt.experimental.pipeline.torch.llm.lora.configs.LoRAUseCaseConfig(*\**, *adapters: list[str]*, *calibration\_dataset: str*, *calibration\_dataset\_path: Optional[str] = None*, *lora\_scaling: list[float]*)

    - Bases: `BaseModel`

Configuration for a LoRA use case (adapter subset + calibration).

- *field* adapters*: list[str]*  *[Required]*

    - - Validated by

    - - `_check_scaling_length`

- *field* calibration\_dataset*: str*  *[Required]*

    - - Validated by

    - - `_check_scaling_length`

- *field* calibration\_dataset\_path*: Optional[str]*  *= None*

    - - Validated by

    - - `_check_scaling_length`

- *field* lora\_scaling*: list[float]*  *[Required]*

    - - Validated by

    - - `_check_scaling_length`

- model\_computed\_fields*: ClassVar[dict[str, ComputedFieldInfo]]*  *= {}*

    - A dictionary of computed field names and their corresponding <cite>ComputedFieldInfo</cite> objects.

- model\_config*: ClassVar[ConfigDict]*  *= {}*

    - Configuration for the model, should be a dictionary conforming to [<cite>ConfigDict</cite>][pydantic.config.ConfigDict].

- model\_fields*: ClassVar[dict[str, FieldInfo]]*  *= {'adapters': FieldInfo(annotation=list[str], required=True), 'calibration\_dataset': FieldInfo(annotation=str, required=True), 'calibration\_dataset\_path': FieldInfo(annotation=Union[str, NoneType], required=False, default=None), 'lora\_scaling': FieldInfo(annotation=list[float], required=True)}*

    - Metadata about the fields defined on the model,
mapping of field names to [<cite>FieldInfo</cite>][pydantic.fields.FieldInfo].

This replaces <cite>Model.__fields__</cite> from Pydantic V1.

## UseCaseRunConfig

- *class* qairt.modules.lora.lora\_config.UseCaseRunConfig(*\*args: Any*, *\*\*kwargs: Any*)

    - Bases: `AISWBaseModel`

Defines the configuration for a specific use case involving one or more LoRA adapters.

- adapters*: List[[AdapterRunConfig](https://docs.qualcomm.com/doc/80-87189-2/topic/qairt-gen-ai-modules-lora.html#qairt.modules.lora.lora_config.AdapterRunConfig)]*

    - A list of LoRA adapter configurations to be used in this use case.
Each adapter is defined by its own <cite>AdapterRunConfig</cite>, specifying parameters
such as adapter name and scaling factor.

- use\_case\_name*: str*

    - A unique identifier for the use case, representing a single adapter or a group of adapters.

## AdapterRunConfig

- *class* qairt.modules.lora.lora\_config.AdapterRunConfig(*\*args: Any*, *\*\*kwargs: Any*)

    - Bases: `AISWBaseModel`

Defines the configuration parameters for executing a LoRA (Low-Rank Adaptation) model.

This configuration is used to control how the LoRA adapter is applied during model inference.

- adapter\_name*: str*

    - The name or identifier of the LoRA adapter to be used during execution.

- alpha*: float*  *= 1.0*

    - A scaling factor applied to the LoRA weights.

Last Published: Aug 26, 2026

[Previous Topic
run\_evaluation()](https://docs.qualcomm.com/bundle/publicresource/80-87189-2/topics/qairt-pipeline-generation.md)