feat/<short-name> either add parameter "extract_image_block_output_dir" in unstructured api call of shared.PartitionParameters or include base46 encoded images in meta data when chunking stratergy ="by_title"

Open

#258 opened on Apr 5, 2025 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Assessment

Difficulty: 5/5
Estimated time: Over a week
Newbie friendliness: 25/100

Issue type: Feature
Clarity: Needs clarification
Activity status: Stale
Tech stack: python
Domain: api

Research direction

The issue does not identify any files, tests, or entry points. Begin by locating the client API handling for chunking_strategy="by_title" and image extraction, then clarify whether completion means returning image_base64 metadata or enabling extract_image_block_output_dir, with tests covering the chosen behavior.

Written by the indexing model from the issue text.

Description

enhancement

Is your feature request related to a problem? Please describe.
. Ex. I'm always frustrated when Yes. I'm currently using the Unstructured API with chunking_strategy="by_title" to segment documents for downstream semantic search and RAG applications. However, when using this chunking strategy, I’m unable to retrieve the Base64-encoded images in the document (i.e., no image_base64 in the metadata of returned elements). Additionally, the parameter "extract_image_block_output_dir"—which allows extracted image blocks to be saved—is unavailable or not functioning when is used. unstructured by api call

This prevents me from building a robust multimodal pipeline that includes both structured text chunks and their corresponding visual content.

Describe the solution you'd like
I’d like one of the following to be supported when using chunking_strategy="by_title" in the Unstructured API:

(Preferred) Embed Base64-encoded image (image_base64) in the metadata of returned elements—just as is supported in r hi-res strategies.

(Alternative) Enable the extract_image_block_output_dir parameter to work unstructured by api call , so that extracted image blocks are saved to a specified directory even during chunked extraction.

Describe alternatives you've considered
A clear and concise description of any alternative solutions or features you've considered.

Additional context
Add any other context or screenshots about the feature request here.

Dominant language: Python
Stars: 119
Forks: 22
Avg merge: 1d 21h
Merged PRs (30d): 1

Contributor guide

Open the contributing guide

First steps

Read the whole issue, then the project's contributing guide.
Comment on the issue to say you are picking it up — it saves two people doing the same work.
Fork the repository and make your change on a branch.
Open a pull request that references the issue number.

Similar issues

issue: imported chats all show as unread in the sidebar Open

bug confirmed issue

Difficulty 2/5 1-3 hours Newbie friendliness 75/100

open-webui/open-webui#30750 · 1 comment · Opened Sep 24, 2026
OGS AmBe example has no source.timing block, so every event sits at time zero Open

Difficulty 2/5 1-3 hours Newbie friendliness 75/100

lanl/ScintiPix#61 · Opened Sep 24, 2026
Make Liquid Glass the default theme for the Research build Open

enhancement

Difficulty 2/5 1-3 hours Newbie friendliness 75/100

OpenwaterHealth/openmotion-bloodflow-app#604 · 1 comment · Opened Sep 24, 2026
tests/puppeteer/test_openai_api_key_ui.js targets controls that do not exist and is run by no workflow Open

Difficulty 2/5 1-3 hours Newbie friendliness 70/100

LearningCircuit/local-deep-research#6766 · Opened Sep 23, 2026
Delete unused args argument from main() in Python files Open

good first issue

Difficulty 1/5 Under an hour Newbie friendliness 90/100

ros2/launch#1014 · Opened Sep 23, 2026

More Python issues