This guide teaches you how to connect OpenAI Hub in WorkBuddy and use a single key to call all AI models.
Why Connect to OpenAI Hub
The built‑in models in WorkBuddy consume points. When you use too many, you have to top up. After connecting OpenAI Hub:
- One key can call all mainstream models such as GPT‑4, Claude, Gemini, etc.
- Pay by actual usage — more cost‑effective than subscription
- No need to worry about running out of points
Getting Started
1. Obtain the OpenAI Hub API Key
Go to openai-hub.com, register an account, create an API key in the console, and copy/save it.
2. Open WorkBuddy’s Model Configuration
Open WorkBuddy, find the settings icon in the lower‑left corner, click it, then select “Model Settings.”
Scroll all the way to the bottom and click the “+ Configure Custom Model” button.
3. Fill in the Configuration Information
In the pop‑up window, fill in as follows:
Provider Selection: Choose “Custom”
API Endpoint: Enter
https://api.openai-hub.com/v1
API Key: Paste the key you just copied (looks like sk-your-api-key)
Model Name: Enter the model you want to use, for example:
gpt-4o– Latest version of GPT‑4claude-3-5-sonnet-20241022– Claude 3.5gemini-2.0-flash-exp– Gemini 2.0
Note: You must enter the model name accurately; otherwise, the call will fail. If unsure, check the list of supported models in the OpenAI Hub documentation.
After filling in, click “OK” to save.
4. Enable the Configured Model
Go back to the model list, find the model you just added, and check it.
This step is easy to miss — if you don’t check it, the model won’t activate, and WorkBuddy will still use its built‑in model to consume your points.
5. Test It Out
Open any chat window, ask WorkBuddy a question, and see if it responds correctly.
If you get an error such as “Invalid API Key” or “Model does not exist,” check:
- Whether the API key was copied completely (starts with
sk-) - Whether the API endpoint has extra spaces or missing letters
- Whether the model name is one supported by OpenAI Hub
FAQs
Q: It still deducts points after configuration?
A: Check whether the model you configured is checked in the model list. If not, it won’t take effect.
Q: What if it says “Request failed”?
A: The API key may have insufficient balance, or the endpoint address may be incorrect. Check your account status in the OpenAI Hub console.
Q: Can multiple models be configured at the same time?
A: Yes. Repeat the steps above, using a different model name each time. You can switch models in the chat window when using them.
Q: How is OpenAI Hub charged?
A: It is usage‑based; different models have different prices. Check the pricing page on the OpenAI Hub website for details.