Issues with Research Functionality and Model Support in Open WebUI v0.5.10

[ITSpecialist111]

I am currently using Open WebUI version 0.5.10 and have encountered the following issues:

Missing Research Activation Option:

In the chat interface, I do not see a radio button or any option to enable the "Research" functionality, as demonstrated in the YouTube video.
Model Compatibility Concerns:

The video suggests the possibility of using Gemini 2.0 for research purposes. However, upon attempting to configure this, I found that Open WebUI currently only supports OpenAI models for this functionality.
Steps to Reproduce:

Install Open WebUI version 0.5.10.
Navigate to the chat interface.
Attempt to locate and enable the "Research" function.
Try to configure Gemini 2.0 as the model for research tasks.
Expected Behavior:

The chat interface should provide an option to enable the "Research" function, allowing for in-depth research capabilities.
It should be possible to configure and utilize models like Gemini 2.0 for research purposes, as indicated in the referenced video.
Actual Behavior:

The option to enable "Research" is not visible or accessible in the chat interface.
Attempts to configure Gemini 2.0 for research tasks are unsuccessful, with support seemingly limited to OpenAI models.
Additional Information:

Open WebUI Version: 0.5.10
Operating System: Windows 11
Browser: Chrome (lastest)
Relevant Plugins/Functions Installed: Native Tool Calling Pipe

The radio button from the chat box in WebUI looks like a tool that doesn't exisit?

YouTube Video Demonstration: Deep Research for FREE: Open-WebUI can already do it (kinda)
Request for Assistance:

I would appreciate guidance on how to enable the "Research" functionality within the chat interface and clarification on the compatibility of models like Gemini 2.0 for research purposes in Open WebUI.

[marcelsamyn]

hey man I added the research tool in the repo! just add it to your tools in open webui https://github.com/iamarcel/open-webui-utils/blob/main/research_tool.py

[ITSpecialist111]

Ah awesome thank you. I did figure it out in the end mixing the two tools that you mentioned :)
You had change to try the Browse-use WebUI yet? Would be cool to do a video on!

[reelsense]

I had a hard time figuring out what he did based on that video. But I figured out you had to use both native_tool_calling_pipe.py and research_tool.py . Which I confused with https://github.com/iamarcel/deep-research-agent because it was mentioned in the context of the research video or the linked one about deep research. The deep-research-agent repository needs a readme and a link back to this repository to clear up confusion.

I made a bash script to one-shot the deployment of Open WebUI + SearXNG with json export enabled, specifically for native_tool_calling_pipe.py and research_tool.py.
https://git.r21.io/primemover/onepush

• Fully Automated Deployment: Runs non-interactively after an initial prompt for configuration details.
• Optional SearXNG Stack:
  • Deploys a private SearXNG instance on a secure Docker network.
  • Exposes it publicly on a separate subdomain with Basic Auth protection.
  • Allows internal access from Open WebUI tools without authentication for fast, secure searching.
  • Supports an optional Brave Search API key for enhanced, ad-free results.
  • Adds a custom JSON engine to SearXNG for advanced integrations.
• Security First:
  • Firewall Aware: Intelligently detects and manages UFW.
• Robust Service Management: Deploys services via Docker on a shared private network, the recommended method.
• Automatic HTTPS: Uses Nginx as a reverse proxy and Certbot to automatically obtain and renew free Let's Encrypt SSL certificates.
• Production-Ready Automation:
  • System Updates: Configures unattended-upgrades.
  • Container Updates: Deploys Watchtower for automatic container updates.
  • Scheduled Reboots: Implements a robust weekly reboot via a cron.d file.

At the end of the readme it explains how to integrate research_tool.py.