Lesson 7: Adding OpenAI API to Open WebUI

[davewaring]

How to Add GPT-4 to Open WebUI (OpenAI API Setup)

Step 1: Create OpenAI Account

• Visit OpenAI’s website
• Sign up for an account if you haven’t already
• Ensure billing information is set up

Step 2: Get Your API Key

1. Navigate to OpenAI’s platform

   • Sign in to your account
   • Click on “dashboard” then “API Keys”
     
     

2. Generate API Key

   • Click “Create new secret key”
     

     

   • Copy the key immediately (it won’t be shown again)

Step 3: Configure Open Web UI

1. Access Open WebUI Admin Panel
   • Open Open Web UI interface
   • Navigate to Admin Panel
   • Go to Settings
   • Select Connections tab

2. Add API Key
   • Locate OpenAI API section
   • Paste your copied API key
   • Click verify to test connection
     
     

How to Use

1. Start a new chat
2. Click model selector
3. Choose either GPT-3.5 or GPT-4o
4. Begin chatting as you would with regular ChatGPT

Note: Keep your API key secure and never share it publicly. API usage will incur charges based on your OpenAI account’s pricing plan.

Next Steps

Continue on to our next lesson on how to use Open WebUI community tools.

[jordanurbs]

I am assuming this comes with the standard compromise in privacy.

Is the main benefit of this having access to ChatGPT in our local interface?

[steve0x00]

I use ChatGPT daily, far more than OpenWebUI. I would like to put everything in one place if there’s a way to firewall ChatGPT off from any personal information. I wonder if I can have my chats “go ask the big guy” when necessary, without sharing any context. Maybe even show me the prompt and payload before it queries ChatGPT.

[TheMrVon]

If you start a new chat then select whatever model, there will be no context to send. Of course if you have a knowledge base selected any RAG hits will be sent to that off-site LLM.