For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /get-started.md).

Python class

ImagePipeline

ImagePipeline

class max.pipelines.lib.generate_images.ImagePipeline(*args, **kwargs)

source

Bases: Protocol

The pipeline surface generate_images() drives.

PixelGenerationPipeline satisfies it.

execute()

execute(inputs)

source

Runs one request through the whole denoising loop.

Parameters:

inputs (PixelGenerationInputs[PixelContext])

Return type:

dict[RequestID, GenerationOutput]

pipeline_config

property pipeline_config: PipelineConfig

source

The config that names the model and its architecture.