[GH-ISSUE #8980] Some feature suggestions (Markitdown as contect extraction engine, more connections for Embedding and ComfyUI, UI tools) #135068

New Issue

2026-05-25T01:38:04-05:00

GiteaMirror commented

2026-05-25 01:38:04 -05:00

Originally created by @YarvixPA on GitHub (Jan 27, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/8980

First of all, I would like to thank and congratulate everyone involved in the development and support of the interface. It’s a great interface, and I would like to share some suggestions that you may want to consider.

Add "Markitdown" as a "Content Extraction Engine", supporting:
- PDF
- PowerPoint
- Word
- Excel
- Images (EXIF metadata and OCR)
- Audio (EXIF metadata and speech transcription)
- HTML
- Text-based formats (CSV, JSON, XML)
- ZIP files (iterates over contents)
Markitdown on GitHub
Allow adding more Ollama servers as "Embedding Model Engine", useful when the main desktop is off but another user needs access (similar to adding more connections).
Add the option to specify which user roles can access a function.
Suggestion for the interface: tools like "Image Generation" and "Web Search" could feature buttons, similar to ChatGPT, Qwen, and Deepseek, making them more practical. Use the plus icon only for attachments.
Allow adding more ComfyUI Base URLs, useful for situations where the main desktop isn't running Ollama but another user needs access.
Add "Canva" with a window like ChatGPT for writing and coding, including buttons for executing and swapping to artifacts based on the code canvas.
Add options to delete uploads and clean or delete embedding vectors, possibly in the "Embedding" window. This would be useful for managing data storage and cleaning up unnecessary data.

Originally created by @YarvixPA on GitHub (Jan 27, 2025). Original GitHub issue: https://github.com/open-webui/open-webui/issues/8980 First of all, I would like to thank and congratulate everyone involved in the development and support of the interface. It’s a great interface, and I would like to share some suggestions that you may want to consider. 1. **Add "Markitdown" as a "Content Extraction Engine"**, supporting: - PDF - PowerPoint - Word - Excel - Images (EXIF metadata and OCR) - Audio (EXIF metadata and speech transcription) - HTML - Text-based formats (CSV, JSON, XML) - ZIP files (iterates over contents) [Markitdown on GitHub](https://github.com/microsoft/markitdown) 2. **Allow adding more Ollama servers as "Embedding Model Engine"**, useful when the main desktop is off but another user needs access (similar to adding more connections). 3. **Add the option to specify which user roles can access a function**. 4. **Suggestion for the interface**: tools like "Image Generation" and "Web Search" could feature buttons, similar to ChatGPT, Qwen, and Deepseek, making them more practical. Use the plus icon only for attachments. 5. **Allow adding more ComfyUI Base URLs**, useful for situations where the main desktop isn't running Ollama but another user needs access. 6. **Add "Canva" with a window like ChatGPT for writing and coding**, including buttons for executing and swapping to artifacts based on the code canvas. 7. **Add options to delete uploads and clean or delete embedding vectors**, possibly in the "Embedding" window. This would be useful for managing data storage and cleaning up unnecessary data.

GiteaMirror closed this issue

2026-05-25 01:38:07 -05:00

Sign in to join this conversation.

Branches Tags

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: github-starred/open-webui#135068