`pixano_inference.pydantic.tasks.multimodal.conditional_generation`

Pydantic models for text-image conditional generation task.

`TextImageConditionalGenerationInput`

Bases: BaseModel

Input for text-image conditional generation.

Attributes:

Name	Type	Description
`prompt`	`str \| list[dict[str, Any]]`	Prompt for the generation. Can be a string or a list of dictionaries to apply a chat template.
`images`	`list[str \| Path] \| None`	Images for the generation. Can be None if images are passed in the prompt.
`max_new_tokens`	`int`	Maximum number of new tokens to generate.
`temperature`	`float`	Temperature for the generation.

`TextImageConditionalGenerationOutput`

Bases: BaseModel

Output for text-image conditional generation.

Attributes:

Name	Type	Description
`generated_text`	`str`	Generated text.
`usage`	`UsageConditionalGeneration`	Usage of the model for the generation.
`generation_config`	`dict[str, Any]`	Configuration for the generation.

`TextImageConditionalGenerationRequest`

Bases: BaseRequest, TextImageConditionalGenerationInput

Request for text-image conditional generation.

`to_input()`

Convert the request to the input.

Source code in pixano_inference/pydantic/tasks/multimodal/conditional_generation.py

def to_input(self) -> TextImageConditionalGenerationInput:
    """Convert the request to the input."""
    return self.to_base_model(TextImageConditionalGenerationInput)

`TextImageConditionalGenerationResponse`

Bases: BaseResponse

Response for text-image conditional generation.

Attributes:

Name	Type	Description
`data`	`TextImageConditionalGenerationOutput`	Output of the generation.

`UsageConditionalGeneration`

Bases: BaseModel

Usage metadata of the model for text-image conditional generation.

Attributes:

Name	Type	Description
`prompt_tokens`	`int`	Number of tokens in the prompt.
`completion_tokens`	`int`	Number of tokens in the completion.
`total_tokens`	`int`	Total number of tokens.