DeepsetFileToBase64Image

Convert image files from disk into base64-encoded images for use in visual question answering pipelines.

Deprecation Notice

This component is deprecated. Use ImageFileToImageContent from Haystack instead. Existing pipelines that use this component continue to work for now.

Key Features

• Reads image files from local disk and converts them to Base64Image objects.
• Supports PNG, JPG, JPEG, and GIF file formats.
• Accepts documents with file_path metadata or direct file sources.
• Configurable image detail level for vision model input sizing.

Configuration

1. Drag the DeepsetFileToBase64Image component onto the canvas from the Component Library.
2. Click the component to open the configuration panel.
3. Set detail to control how the image is resized before encoding.

Connections

DeepsetFileToBase64Image accepts documents with file_path metadata or direct sources as input. It outputs documents and base64_images.

Connect DeepsetFileDownloader to provide documents with local file paths. Connect the base64_images output to a visual Generator such as DeepsetOpenAIVisionGenerator or Base64ImageJoiner.

Usage Example

This example converts downloaded image files to base64 for a vision Generator:

components:
  image_downloader:
    type: deepset_cloud_custom_nodes.augmenters.deepset_file_downloader.DeepsetFileDownloader
    init_parameters:
      file_extensions:
        - .png
        - .jpg
        - .jpeg
      sources_target_type: str
      max_cache_size: 100

  file_to_image:
    type: deepset_cloud_custom_nodes.converters.file_to_image.DeepsetFileToBase64Image
    init_parameters:
      detail: auto

connections:
  - sender: image_downloader.documents
    receiver: file_to_image.documents

Parameters

Inputs

Parameter	Type	Default	Description
documents	Optional[List[Document]]	None	Documents with file_path in metadata pointing to image files.
sources	Optional[List[Union[str, Path, ByteStream]]]	None	Direct file paths or byte streams to convert.

Outputs

Parameter	Type	Default	Description
documents	List[Document]		Documents that had convertible image files.
base64_images	List[Base64Image]		Base64-encoded images ready for vision Generators.

Init Parameters

These are the parameters you can configure in Pipeline Builder:

Parameter	Type	Default	Description
detail	Literal["auto", "low", "high"]	auto	Controls how the image is resized before encoding. See the OpenAI vision documentation for details.

Run Method Parameters

These are the parameters you can configure for the component's run() method. This means you can pass these parameters at query time through the API, in Playground, or when running a job. For details, see Modify Pipeline Parameters at Query Time.

Parameter	Type	Default	Description
documents	Optional[List[Document]]	None	Documents with file_path in metadata.
sources	Optional[List[Union[str, Path, ByteStream]]]	None	Direct file paths or byte streams to convert.