[PR #99] [MERGED] Ollama Dynamic Vision Pipeline #304

New Issue

2026-02-15T20:15:01-05:00

yindo commented

2026-02-15 20:15:01 -05:00

📋 Pull Request Information

Original PR: https://github.com/open-webui/pipelines/pull/99
Author: @atgehrhardt
Created: 6/18/2024
Status: ✅ Merged
Merged: 6/18/2024
Merged by: @tjbck

Base: main ← Head: main

📝 Commits (4)

58d64c4 Pipeline that automatically extracts image, passes to an ollama vision model, and overrides the response. This allows for you to have a primary text model and temporarily switch to a vision model whenever there is an image attached
cc93657 Adding metadata
7d817b6 Fixed metadata. Fixed image retention.
c360d2b Cleaning up

📊 Changes

1 file changed (+91 additions, -0 deletions)

View changed files

➕ examples/filters/dynamic_ollama_vision_filter_pipeline.py (+91 -0)

📄 Description

This allows you so select an Ollama model to override with another Ollama image model. If an image is attached, it will send the image over to the override model and that model will return it's description and the initial model will take back over.

This isn't perfect, but allows you to take any text model, check the vision option, and have it process images without breaking the UI by leaning on another model.

This is JUST A STARTING POINT.

There is a LOT more we can do to make this more robust, but it serves as a good starting point.

_{🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.}

## 📋 Pull Request Information **Original PR:** https://github.com/open-webui/pipelines/pull/99 **Author:** [@atgehrhardt](https://github.com/atgehrhardt) **Created:** 6/18/2024 **Status:** ✅ Merged **Merged:** 6/18/2024 **Merged by:** [@tjbck](https://github.com/tjbck) **Base:** `main` ← **Head:** `main` --- ### 📝 Commits (4) - [`58d64c4`](https://github.com/open-webui/pipelines/commit/58d64c4fcec2c6a20f71e645cc14014b68b24b39) Pipeline that automatically extracts image, passes to an ollama vision model, and overrides the response. This allows for you to have a primary text model and temporarily switch to a vision model whenever there is an image attached - [`cc93657`](https://github.com/open-webui/pipelines/commit/cc936574e5f7fba77aadc8e8f8180751f3251eb9) Adding metadata - [`7d817b6`](https://github.com/open-webui/pipelines/commit/7d817b6b6abe7ffbc7b7bdbde1f52a6666476943) Fixed metadata. Fixed image retention. - [`c360d2b`](https://github.com/open-webui/pipelines/commit/c360d2bda8fc518d2ba5c7be5b51ffbe1dede4cb) Cleaning up ### 📊 Changes **1 file changed** (+91 additions, -0 deletions) <details> <summary>View changed files</summary> ➕ `examples/filters/dynamic_ollama_vision_filter_pipeline.py` (+91 -0) </details> ### 📄 Description This allows you so select an Ollama model to override with another Ollama image model. If an image is attached, it will send the image over to the override model and that model will return it's description and the initial model will take back over. This isn't perfect, but allows you to take any text model, check the vision option, and have it process images without breaking the UI by leaning on another model. This is JUST A STARTING POINT. There is a LOT more we can do to make this more robust, but it serves as a good starting point. --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>

yindo added the pull-request label 2026-02-15 20:15:01 -05:00

yindo closed this issue

2026-02-15 20:15:01 -05:00

yindo referenced this issue

2026-02-15 20:15:18 -05:00

[PR #304] [MERGED] Add examples/pipelines/integrations for DifyAI and N8N #371

Sign in to join this conversation.

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: open-webui/pipelines#304